No Section 8 Recovery for Being Kept out of a Market Which the Generic Had No Right to Enter

AstraZeneca Canada Inc v Apotex Inc 2017 FC 726 Barnes J [Omeprazole Accounting]
            1,292,693 / omeprazole formulation / LOSEC

Until recently amended, the NOC proceedings and infringement actions were separate, and a declaration of invalidity in an NOC proceeding did not have in rem effect. This raised the possibility that the generic might prevail in the NOC proceeding, and lose in the subsequent action. If the generic prevailed in the NOC proceeding, it would be entitled to damages under s 8 for having been kept out of the market by the statutory stay on the issuance of the NOC. But if it lost in the subsequent infringement action, this means the patent is and always was valid, and it seems wrong to award the generic damages for being kept out of a market which it had no right to enter.

Now that the NOC Regulations have been amended to turn the NOC proceedings into an action, this particular problem will no longer arise, but a closely related problem remains. What if the product is covered by two patents, and the generic wins the NOC proceeding/ action in respect of one patent, but loses in respect of the other? Again, the generic will be entitled to s 8 damages for being kept out of a market which it had no right to enter. That problem arose in this case. Both the 2,133,762 patent and the ‘693 patent were listed on the Patent Register against LOSEC. Apotex prevailed in NOC proceedings related to the 2,133,762 patent (2004 FC 313, holding infringement was not established), but then lost in respect of the ‘693 patent, which gave rise to the litigation in this case.

The problem had earlier been addressed in respect of this very patent in the s 8 proceedings arising from the ‘762 patent: 2012 FC 559 Hughes J (here) aff’d 2013 FCA 77 (here) motion for reconsideration dismissed, 2016 FCA 194. AstraZeneca raised the relevance of the then ongoing infringement action in respect of the ‘693 patent, asking for a stay of the s 8 proceedings pending the ‘693 action. Hughes J refused, saying the following [148], in a passage affirmed and quoted on appeal [6]:

A Court hearing the pending infringement action, if it concludes that the patent is valid and has been infringed by Apotex in making the omeprazole drug that is the subject of these proceedings, can at that time craft a remedy that is appropriate, having in mind any compensation awarded in these proceedings.

(See similarly 2011 FCA 364 (blogged here).) Since Barnes J is now that very “Court hearing the . . . infringement action,” Barnes J was on firm ground in concluding that:

[214] What I take from the above-noted statements and particularly those of the Federal Court of Appeal is that, as the section 8 reference Judge, I have the discretion to take into account the intervening infringement finding, among other relevant facts, and to craft an appropriate remedy.

The question then is how to properly exercise that discretion. The FCA had dropped some strong hints. In 2013 FCA 77 [7] the Court stated:

It will be for the judge trying the infringement action to ensure that overall, taking both proceedings together, a party is compensated for its provable loss, if any, on proper principles, no more and no less.

The FCA repeated this in denying the motion for reconsideration: 2016 FCA 194, [24]. Similarly, in the Lovastatin decision, 2011 FCA 364 (blogged here) the FCA held:

[37] [S]ubsection 8(5) confers a broad discretion on the court when assessing the amount of compensation that the second person must pay. It provides that the court “shall take into account all matters that it considers relevant to the assessment of the amount,” including any conduct by either party that contributed to the delay in the disposition of the first person’s application for prohibition. In my view, this provision enables the Court to determine in its discretion whether, and to what extent, a second person’s claim for compensation should be reduced, or eliminated.

On the facts of this case, Barnes J notes that :

[217] What I am left with is a situation where, in order to recover its “losses” from being barred from selling Apo-Omeprazole between January 3, 2002 to December 30, 2003 in the face of AstraZeneca’s 762 Patent, Apotex necessarily had to infringe AstraZeneca’s 693 Patent.

In other words, Apotex was seeking damages for having been kept out of a market that it had no right to enter. Unsurprisingly, Barnes J concluded that:

[219] It follows that Apotex is not entitled to recover under section 8 of the NOC Regulations because it suffered no loss by being kept out of the marketplace between January 3, 2002 and December 30, 2003.

I believe this is the first decision to directly so hold, but this conclusion was strongly foreshadowed in the prior decisions, and I don’t doubt that it will be upheld on appeal and followed in the future.

Does an NIA Need to Be Proven on the Balance of Probabilities

AstraZeneca Canada Inc v Apotex Inc 2017 FC 726 Barnes J [Omeprazole Accounting]
            1,292,693 / omeprazole formulation / LOSEC

Yesterday’s post provided an overview this decision, and dealt with the non-infringing alternatives developed for litigation. Apotex also proposed that it could have sourced non-infringing product from two third parties, Estevé and Kudco.

Kudco presents an important issue. Kudco held the US rights a non-infringing omeprazole formulation under license from a French company. Kudco executives testified that they would have had the authority to supply product to Apotex for sale in Canada [174], and that they would have been willing to do so if a deal could be reached [177]. Nonetheless, Barnes J did not accept this as an available alternative, primarily because, on the facts, he held that it was more likely than not that a deal would not have been reached between the parties, because Kudco would have asked more than Dr. Sherman would have been willing to pay [191].¹ Barnes J therefore held (my emphasis):

[192] I am not, therefore, satisfied that Apotex is more likely than not to have entered into some form of licensing agreement with Kudco during the infringing period. If I am wrong about the standard of proof that applies such that the test in Athey v Leonati, [1996] 3 SCR 458, 140 DLR (4th) 235, is applicable, I would fix the possibility of reaching an agreement with Kudco to supply the Canadian market by the beginning of the infringing period at 15% and at a royalty rate of 35% on Apotex’s net sales.

In both Lovastatin [74] and Venlafaxine 2016 FCA 161 [56], the FCA held that the burden lies on the defendant to establish the factual relevance of a non-infringing alternative on the balance of probabilities. In both cases the FCA cited as authority Rainbow Caterers [1991] 3 SCR 3, 14, which does indeed say that. The problem, as I have discussed, is that Athey v Leonati [1996] 3 SCR 458, [27] says “Hypothetical events . . . need not be proven on a balance of probabilities. Instead, they are simply given weight according to their relative likelihood” (my emphasis). It is not clear to me whether these two holdings can be reconciled. Moreover, the point was thinly reasoned in Rainbow Caterers, and fully reasoned in Athey v Leonati, and the relative likelihood approach appears to be accepted and routinely applied in most other areas of law, eg personal injury.

Because Barnes J made a specific finding as to the relative likelihood of the hypothetical event for the Kudco NIA, and because Apotex failed in respect of all the other NIAs, if there is an appeal, this squarely raises the question of whether the FCA should revisit this question in light of the broader SCC jurisprudence. I am not criticizing the FCA for having followed Rainbow Caterers in the first place. It is SCC authority which does stand for that proposition on this point, and moreover, the point was conceded by Apotex in oral argument: Lovastatin [74]. But on fuller consideration, Rainbow Caterers may not be the best authority.

It seems the issue really only arises in respect of the Kudco NIA. On the relative likelihood approach, a hypothetical possibility “will be taken into consideration as long as it is a real and substantial possibility and not mere speculation”: Athey [27]. It is clear that the sourcing the product from Estevé would not meet this threshold. Barnes J held that the Estevé product was not an available NIA because Estevé had entered into an exclusive North American licence with Mylan, and Mylan would have had to waive its rights for Estevé to supply Apotex. On the facts. Barnes J held that “[t]he likelihood that Mylan would have waived its Canadian rights in favour of a significant competitor seems very remote” [169]. With respect to the post-infringement alternatives developed for litigation, Barnes J expressly applied on the balance of probabilities test, but my sense from is discussion is that those alternatives did not rise beyond speculation.

1 It is contrary to economic logic to suppose that the parties would have left money on the table by failing to make a deal, and Barnes J remarked that “Evidence in the form of an economic model would have been more persuasive than this kind of hypothetical anecdotalism: [192].

NIA Developed Solely for Litigation

AstraZeneca Canada Inc v Apotex Inc 2017 FC 726 Barnes J [Omeprazole Accounting]
            1,292,693 / omeprazole formulation / LOSEC

In the accounting of profits portion of this bifurcated trial,¹ the parties settled most issues, but required determination of four specific points. The first, discussed in this post, is whether Apotex had a non-infringing alternative available during the period of infringement. The decision raises two significant issues. One is whether an infringer must know at the time of the infringement that the NIA is non-infringing, for it to be considered available. More generally, this case is novel in that Apotex attempted to establish case an NIA “defence”² based solely on alternatives which had been developed only for the purposes of litigation [24], [92]. This contrasts with the leading cases to date, Lovastatin FCA 2015 FCA 171 and Perindopril FCA 2017 FCA 23. In Lovastatin (see here) the putative NIA had actually been developed and commercialized by Apotex prior to the infringement; in Perindopril (here) the issue involved export of infringing product made in Canada, and the NIA was that the product would have been exported to the same countries, and made by the same process in it other jurisdictions where it was not patented. In the leading US decision, Grain Processing 185 F.3d 1341 (Fed. Cir. 1999), the alternative did not actually exist at the date of infringement, but was commercialized almost immediately after infringement was established. In this case, in contrast, the putative NIAs were seven alternative formulations which Apotex never commercialized and has no intention of ever commercializing [24]. They were developed solely for the purpose of arguing that they could and would have been commercialized in the but for world.

Apotex won on the law, but lost on the facts. Barnes J held that these types of alternatives could in principle be considered, but he found that Apotex failed to establish that the particular alternatives it developed would have been commercializable, or that they would have received regulatory approval. While this ultimately turned on the specific facts of the case, there is perhaps a broader lesson. The testing of the alternative formulations that was done in this case fell far short of what would be required for regulatory approval. The tests used to predict both bioequivalency and stability were appropriate for use in a formulation screening exercise, but this was far from enough to persuade Barnes J that the formulations would probably have been approved and commercially successful [139], [146]. This suggests that while post-infringement NIAs are acceptable in principle, it will be difficult for a generic to surmount the necessary evidentiary hurdles to establish the NIA on the facts. Apotex also failed to establish it would have sourced non-infringing product from a third party (which I will discuss tomorrow) [196]. On the whole, this decision should give comfort to patentees who were concerned that the acceptance of the NIA defence would open the floodgates to speculative defences.

On the law, AstraZeneca made three arguments of principle. First, relying on MacKay J’s decision in Wellcome v Apotex (TMP) (1998), 82 CPR (3d) 466, [32]-[33}, it argued that a valid NIA must be perceived or known by the infringer to be non-infringing at the point of the infringement [12]. [13]. To simplify slightly, in TMP the final product, TMP, could be made by two processes, both of which were patented by Wellcome. At trial, one of the process patents was held to be valid and infringed, but the other was held to be invalid. Apotex argued that it – or more precisely, its suppliers – could have used the process protected by the invalid patent. MacKay J rejected this argument, saying “there is really no evidence that at the time the product was acquired there was available product then known to be non-infringing” [33]. In part this was because the alternative patent was not then known to be invalid, and in part because Apotex did not know which process was used by its suppliers. On the whole, though the discussion is brief and somewhat cryptic, TMP does seem to me to support AstraZeneca’s argument. Barnes J distinguished TMP on the basis that MacKay J also considered whether Apotex “could have known” [14]. I don’t find this distinction entirely persuasive. MacKay J’s point, as I read it, is that not only did Apotex not actually know, it might not have been possible for it to know – this reinforces the importance of knowledge, rather than diminishing it. In any event, Barnes J did accept that knowledge is relevant, saying “[k]nowing whether or not a proposed NIA would infringe is, of course, a factor in determining whether the infringer ‘would have’ employed it in place of the infringing product” [14], though “this falls well short of making prior knowledge of non-infringement an absolute pre-requisite to the assertion of a NIA” [13]. Thus, Barnes J agreed with MacKay J that knowledge that the NIA is non-infringing is relevant, and the difference between them, if any, is whether it should be a complete bar.

But why should it be relevant at all? As Barnes J noted, it apparently follows from the “would have” branch of the test for availability, evidently on the notion that the infringer – eg Apotex in TMP – would not have known to insist that its suppliers use the non-infringing alternative because it did not know that it was non-infringing. But this seems dubious in principle, as this approach allows an invalid patent to shield behind a valid one. To go back to first principles, the reason alternatives are considered is that the return to a patentee should be commensurate with the value of the patented technology, which is equal to its value as compared with the best non-infringing alternative: Lovastatin FCA [56]. A technology may be innovative and patentable, yet not valuable, if the product is no better or cheaper than a mundane alternative. If the infringer must know that a patent is invalid before that technology can be used as an alternative, the patentee can improperly inflate the apparent value of the innovative technology by patenting obvious technologies that serve the same end, even though those patents are ultimately invalidated. This is a kind of remedial evergreening that is not addressed by the usual expedient of invalidating the offending patents. It will be interesting to see how the law continues to develop on this point.

AstraZeneca also argued that an NIA “must be ‘foreseeable’ to the infringer at the relevant time” [13], citing Lovastatin FCA [93]-[94]. Barnes J dismissed this as a “stray reference” [15], saying

If foreseeability meant that the infringer must have the asserted NIA in mind at the time of the infringement, it could potentially punish those who had no idea their product was infringing while rewarding those who had an appreciation of the risk and courted it, but nevertheless had a back-up, work-around solution available.

Barnes J’s observation strikes me as compelling.

Finally, citing Grain Processing 185 F 3d 1341 (Fed Cir 1999) and Micro-Chemical 318 F 3d 1119 (Fed Cir 2003), AstraZeneca argued that an NIA “requiring the infringer to ‘invent around the patented technology’ is not considered to be ‘available’ to the infringer” [16]. Barnes J rejected this, saying that cases stand only for the proposition that “[t]he time and effort of coming up with a non-infringing solution is certainly relevant to whether the infringer would have pursued it, but they are not absolute barriers to the defence” [16]. That conclusion strikes me as both a correct reading of those cases, and sound as a matter of principle.

In conclusion, Barnes J held that as a matter of law:

in the hypothetical, but for pharmaceutical world the infringer’s failure to produce a viable NIA formulation in the real world is not a threshold bar to the use of the NIA defence . [18]

The correct question is simply “Could the infringer have made the product had it attempted to do so at the relevant time and would the infringer have sold the product on some reasonable financial basis in substitution for the infringing product?” [18].

While the law was largely decided in Apotex’ favour, Barnes J pointed out that post-infringement development of an NIA may present serious problems of proof, which were manifest in this case [20].

One general point is that the Court should be wary of hindsight bias when assessing the availability of post-infringement development of an NIA [21], in the same way as the Court must be wary of hindsight bias was assessing an obviousness attack.

But this case did not turn on hindsight bias. It turned on more basic evidentiary problems:

[24] When a pharmaceutical NIA has been created and has obtained regulatory approval, one is not left to wonder whether it “could” have been available for use (assuming a capacity to obtain it in commercial amounts). In this case, however, Apotex’s self-created NIAs were made in non-commercial batches, without full stability, bioequivalency or clinical studies, and without obtaining the required regulatory approvals for commercial use.

Nontheless, Barnes J did not dismiss Apotex’ tests out of hand, but at the end of the day, the data presented by Apotex was flawed and speculative in many respects: “shortcuts like the ones employed by Apotex may be acceptable in the early stages of product selection, but they are entirely unacceptable in the context of proving the viability of a NIA for the purpose of this case” [143]. For much the same reasons, Apotex could not establish that its proposed alternatives would have received regulatory approval [144]-[152].

1 The liability portion is Apotex Inc v AstraZeneca Canada Inc 2017 FCA 9 var’g 2015 FC 322 as amended by 2015 FC 671: see FCA Claim Construction FCA Overbreadth FC Product Claim FC Foreign Issue Estoppel

2 I don’t like the term NIA “defence” because in most circumstances it does not serve as a complete defence, but only to reduce the quantum that would otherwise be awarded. Nonetheless, the term is convenient, and it seems to have become established, so I will use it, albeit with this caveat.

Companion Case to Pfizer v Apotex / ODV

Pfizer Canada Inc v Teva Canada Ltd 2017 FC 777 Brown J

2,436,668 / desvenlafaxine (ODV) / PRISTIQ / NOC

This is a companion case to Pfizer Canada Inc v Apotex Inc 2017 FC 774, blogged here on the obvious-to-try issue. The patent was the same, and the invention story is the same, and the discussion of the facts is largely verbatim repetition [4]. Unlike Apotex, Teva did not argue non-infringement, anticipation or double patenting, but instead restricted itself to obviousness and inutility challenges, which were also raised by Apotex [5]. The discussion of the law on those points is largely verbatim repetition, with a few differences between the decisions to address points made by one party but not the other. While I admit that I did not re-read the entire Teva decision, the differences do not appear to warrant a separate blog post on this case.

Pre-issuance Defects in Administrative Process Cannot Render a Patent Void

Apotex Inc v Pfizer Inc 2017 FCA 201 Gauthier JA: Stratas, Boivin JJA aff’g 2016 FC 136 Diner J
            1,339,132 / latanoprost / XALATAN

The precise issue in this case will probably never arise again, as it is “is the first and most likely the last case involving section 73 of the 1989 Act” [4]. Nonetheless, Gauthier JA’s decision for the Court sets out an important general principle that “pre-patent issuance defects in the administrative process for applying for a patent cannot be relied upon by an alleged infringer to render a patent void” [59]. The Court also suggested a principle of interpretation which may have even more far-reaching implications, to the effect that the courts should strive to interpret the Canadian Patent Act so as to achieve substantive consistency with the patent law of our trading partners.

The applicant that filed the application which matured into the ‘132 patent (Pharmacia Aktiebolag, Pfizer’s predecessor) was a large entity. It did not claim small entity status, and it paid large entity fees — except for the final fee, for which, for reasons which are unknown, it paid the small entity fee [8]. For reasons which are also unknown, the Patent Office accepted that fee, even though small entity status had not been claimed [9]. On subsequent inquiry by the patent agent, the Patent Office  acknowledged, erroneously, that large entity fees had been received [10]. Thus, the applicant was a large entity, and the proper large entity fee had never been paid [11].

On these facts, in the context of s 8 litigation, Apotex sought a motion for summary judgement that the 132 patent is invalid for failure to pay the proper application fee [12]. Diner J dismissed the motion, and Apotex appealed.

Apotex made a straightforward argument [49]. It relied on s 59, which provides that when defending an infringement action, a defendant “may plead […] any fact or default which by this Act or by law renders the patent void,” in combination with subsection 27(1) of the 1989 Act, which provided (my emphasis):

inventor […] may, on presentation to the Commissioner of a petition setting out the facts […] and on compliance with all other requirements of this Act, obtain a patent granting to him an exclusive property in the invention

27(2) The requirement to pay the fee is the in Act, so the applicant had not complied with all other requirements of the Act, and this, the argument went, renders the patent void. The wording of the current 27(1) is slightly different, but the same argument would arise, as it provides that the Commissioner shall issue a patent if “all other requirements for the issuance of a patent under this Act are met.”

Gauthier JA remarked that the “literal approach” advocated by Apotex was attractive, but she rejected it in light of the context and purpose [50]-[51]. The FCA held that it was bound by Fada Radio [1927] SCR 520, in which the SCC had held that an untrue statement in the application (that the invention had not been patented with the applicant’s knowledge in any country) would not invalidate a granted patent, which stands for this “essential concept”:

[57] pre-patent issuance defects in the administrative process for applying for a patent cannot be relied upon by an alleged infringer to render a patent void.

Similarly, the FCA cited its own prior caselaw for the principle that

[58] defects in the pre-patent issuance process that do not come within the ambit of provisions dealing expressly with the voidance of a patent, like section 53 of the 1989 Act, cannot be relied upon by an alleged infringer to render a patent void.

This is evidently a general principle, not tied to the 1989 version of 27(1).

Gauthier JA also referred to the absurdity of “enabling an alleged infringer to void a patent (here a successful pharmaceutical patent worth millions if not billions of dollars) say ten years later or even after its expiration, on the basis that the petitioner was a few pennies short” [72].

Therefore, “the law” referred to by s 59 is substantive patent law, not administrative law matters which might be raised on judicial review: [69], [70]. Gauthier JA noted that the Economic Action Plan Act, s 138, not yet in force, “makes it clear that non-payment of fees payable before the issuance of a patent will not invalidate the said patent” [3], but this played no part in her reasoning, so the interpretation stands even if this provision is never proclaimed in force.

This seems to me entirely sound as a purposive analysis of the Act. It also strikes me as consistent with the literal text of the Act. S 59 refers to any fact which renders the patent void, and nothing in 27(1), then or now, refers to validity; it refers to the conditions on which the Commissioner may (or, now, shall) issue a patent.

The Court’s final observation may have even more wide-ranging implications:

[77] Considering the importance of patents nowadays, and the importance given to intellectual property law in trade treaties, courts should obviously be careful before adopting an interpretation that would put Canada at odds with its trading partners. Thus, I am comforted by the fact that my purposive interpretation of the 1989 Act does not require the addition or the recognition of the new grounds of invalidity that Apotex’s view in respect of sections 27 and 59 would entail and that could be in direct conflict with those generally recognized in England, Europe and the United States.

This observation is entirely salutary. It is consistent with the observation of the House of Lords as to the importance of establishing a common approach in Europe (see e.g. HGS v Lilly [2011] UKSC 51, noting that the Bundesgerictshof is of the same view). Of course, in Europe the common basis of the EPC provides an additional rationale for harmonization, but Lord Neuberger [96]-[99] also remarked on the importance of more general harmonization in promoting the basic objective of the patent system, to provide an incentive to innovation.

The UKHL has adopted harmonization with the EPO has a principle of interpretation of the UK Act. The FCA in this decision has not gone so far. Gauthier JA simply “drew comfort” from the observation, and it did not play any direct part in her reasoning. Nonetheless, it opens the door to invoking the desirability of harmonization as an independent principle of interpretation of the Patent Act.

Is it "More or less Self-evident That What Is Being Tested Ought to Work"?

Pfizer Canada Inc v Apotex Inc 2017 FC 774 Brown J

2,436,668 / desvenlafaxine (ODV) / PRISTIQ / NOC

Bristol-Myers Squibb Canada Co v Apotex Inc 2017 FCA 190 [Dasatinib FCA] Gleason JA: Webb, Near JJA var’g 2017 FC 296 [Dasatinib FC] Manson J
            2,519,898 / dasatinib / SPRYCEL / NOC

Desvenlafaxine and Dasatinib both apply the “obvious to try” test – endorsed in Sanofi 2008 SCC 61, as “one factor to assist in the obviousness inquiry” [64] – but to opposite effect, with Brown J holding the claimed invention to be non-obvious, while Manson J, now affirmed by the FCA, holding the invention at issue to be obvious. In my view, both decisions applied the test correctly, given the facts as found. And indeed, the FC has almost always applied the test correctly in the years since Sanofi was decided. Perhaps it should not be surprising that the courts apply the law correctly. But I do find it surprising, because there is continuing doctrinal confusion as to the nature of the test, stemming ultimately from Sanofi itself. That confusion has surfaced before, and has now surfaced again in both these recent decisions.

In Sanofi, the SCC stated that:

[65] I am of the opinion that the “obvious to try” test will work only where it is very plain or, to use the words of Jacob L.J., more or less self-evident that what is being tested ought to work.

This lends itself to the interpretation that it must be self-evident, prior to any experimentation, that the claimed invention will solve the problem at hand. But that is difficult to reconcile with the Court’s observation that the obvious to try test is appropriate precisely in “areas of endeavour where advances are often won by experimentation” (Sanofi [68]), which implies that the test applies particularly when it is not self-evident, prior to any experimentation, that the claimed invention will solve the problem at hand.

"The Ability to Inhibit a Biological Target Implicated in Disease" Establishes Utility

Bristol-Myers Squibb Canada Co v Apotex Inc 2017 FCA 190 [Dasatinib FCA] Gleason JA: Webb, Near JJA var’g 2017 FC 296 [Dasatinib FC] Manson J
            2,366,932 / 2,519,898 / dasatinib / SPRYCEL / NOC
            FC double patenting / FC promise

Dasatinib FCA is the first decision of the FCA on utility following the SCC’s Esomeprazole 2017 SCC 36 decision, which “abolished the promise doctrine” [35]. The FCA’s holdings flow straightforwardly from Esomeprazole. This is not surprising, since the scintilla requirement, which is now the only remaining branch of the utility requirement, is of long standing, so it is unnecessary to create new law in the aftermath of Esomeprazole. With that said, pharmaceutical patent litigation only really took off in Canada after compulsory licensing was abolished in 1993, and the promise doctrine has been the primary battleground in utility attacks for much of the period since, so there is little case law on what exactly constitutes a “scintilla” of utility in the pharmaceutical context. The FCA’s statement that “Establishing that a compound has the ability to inhibit a biological target implicated in disease is doubtlessly a useful discovery” [40], is important guidance on this issue.

Both patents at issue relate to dasatinib, which is used to treat chronic myelogenous leukemia (CML). The claim at issue from the ‘932 patent is to the compound dasatinib as such [22]. Dasatinib FC was decided prior to Esomeprazole, and, applying the promise doctrine, Manson J held that the patent promised, inter alia, that dasatinib would be therapeutically useful [31], and that was the standard against which utility was to be assessed. As discussed here, it was undisputed that therapeutic efficacy was not demonstrated or soundly predicted as of the claim date, and the holding that the ‘932 patent lacked utility followed directly [34]. (The ‘898 patent was also held invalid for different reasons, but this was a rare NOC proceeding which was not moot on appeal because the Minister has not yet issued the NOC to Apotex [6]. I’ll discuss the issues raised by ‘898 patent in a subsequent post.)

With the promise doctrine abolished, Manson J’s holding on this point could not stand [8]. The main issue on appeal was therefore whether the claimed invention was useful in light of Esomeprazole.

The SCC in Esomeprazole summarized the correct approach as follows [35]:

[54] First, courts must identify the subject-matter of the invention as claimed in the patent. Second, courts must ask whether that subject-matter is useful — is it capable of a practical purpose (i.e. an actual result)?

[55] The Act does not prescribe the degree or quantum of usefulness required, or that every potential use be realized — a scintilla of utility will do. A single use related to the nature of the subject-matter is sufficient, and the utility must be established by either demonstration or sound prediction as of the filing date (AZT, at para. 56).

On the first point, the FCA held that

[37] [C]ontrary to what Apotex asserts in its supplemental written submissions, the subject-matter of claim 27 of the 932 patent is not the potential therapeutic uses for dasatinib. Rather, the subject-matter of claim 27 is merely the compound, dasatinib, itself. This is all that claim 27 claims, and it is erroneous to expand the subject-matter of the claim beyond what it says.

This is a straightforward reading of Esomeprazole: it is “the invention as claimed” that must be useful, just as it is the claimed invention which must be new and non-obvious. A contrary holding would have re-established the promise doctrine through the back door, by making properties – those disclosed or promised in the disclosure, no doubt – the measure of utility.

At the second stage, the question is whether dasatinib had “at least a scintilla of utility” as of the filing date. But how much is a scintilla? It is a low bar, as the word indicates, but it is more than zero.

On the facts, it was conceded that BMS had established that dasatinib acted in vitro to inhibit a certain class of cellular enzymes, specifically “Src-family PTKs” [39]. It was also known that abnormal activity of PTKs was linked to a variety of diseases, including CML [Dasatinib FC 200]. Apotex argued that this was not enough to establish the requisite scintilla of utility. The FCA disagreed (my emphasis):

[40] Establishing that a compound has the ability to inhibit a biological target implicated in disease is doubtlessly a useful discovery. Here, it was known as of the relevant date that enhanced activity of PTK was involved in many diseases, as stated in the specification and confirmed in the evidence of several of the experts. Thus, discovery of a substance that acted to inhibit certain PTKs represented an important advance and certainly meets the minimal utility requirements that are now applicable following the decision of the Supreme Court in Esomeprazole.

While it is well-established that a “scintilla” of utility is the minimum that is required, the underlined sentence provides important guidance as to what exactly this means in the context of pharmaceuticals. The FCA remarked “parenthetically” that “a similar sort of discovery was found to satisfy the utility requirement in Esomeprazole and Teva Canada Limited v. Novartis AG, 2013 FC 141 [imatinib].” I’m not sure Esomeprazole is helpful. Esomeprazole is the (-) enantiomer of omeprazole, and omeprazole was known to be therapeutically effective in the treatment of gastric acid-related diseases. Moreover, it was known that the active inhibitory agent was an achiral metabolite of omeprazole, so from that it could be very confidently predicted that esomeprazole, which would be transformed into the very same metabolite, would be therapeutically effective: see Esomeprazole FC 2014 FC 638, [165]; and see here. Thus, in Esomeprazole therapeutic utility itself could be soundly predicted, which was not the case for dasatinib. Imatinib, on the other hand, is indeed a close parallel, in my view: see here. I’ll also note that in Moxifloxacin 2014 FC 462 Phelan J held that “good in vitro activity” was sufficient to establish utility. There are also a few other cases with statements indicating a similar standard, those most are strictly distinguishable one way or the other. On the other hand, I am not aware of any cases at all indicating that a “scintilla” of utility is higher. Thus, in my view the FCA’s statement is consistent with the prior cases. But because the promise doctrine has so dominated the law of utility since pharmaceutical litigation took off with the abolition of compulsory licensing, there is remarkably little case law directly on this point. The FCA’s guidance is therefore particularly welcome.

While the FCA did not mention it, this holding is also consistent with the decision of the UKSC in HGS v Lilly [2011] UKSC 51, in which the UK Supreme Court held that the requirement of industrial applicability, the European equivalent of utility, was satisfied when the claimed compound was a member of a family of compounds which were known to have a role in immune and inflammatory responses: see here; and Siebrasse, HGS v Lilly: How Soon Is Too Soon to Patent? (2011) 24 IPJ 41-52. HGS v Lilly is of course not binding on Canadian courts, but as Lord Neuberger discussed [96]-[102], international consistency is very desirable. While patent law is national, the patent incentive is international. Patentees rely on expected profits from multiple jurisdictions when deciding whether it is worthwhile pursuing a particular avenue of research, and an inventor should not have to give up her patent in one jurisdiction to get it in another. Seeking consistency with other jurisdictions, to the extent that our law and sound policy permits, is an important policy consideration. In large part for that reason the UKSC followed the lead of the EPO in setting the standard for industrial applicability. Having taken a step towards harmonization with the rejection of the promise doctrine in Esomeprazole, it is welcome to see that we are not departing from it in the definition of a “scintilla” of utility.

The exact threshold will likely be further refined in subsequent cases. The FCA’s statement that “the ability to inhibit a biological target implicated in disease” establishes utility, has to be read in context. As I understand it, and as indicated by the FCA [40], the link between enhanced activity of PTK and disease was well established: see [Dasatinib FC 200]. There is room to argue that proven in vitro activity against target does not establish utility if the target itself is not as firmly “implicated” in disease. On other hand, the utility accepted in HGS v Lilly was arguably a lower standard, the compound was a member of a family known to have a role in immune and inflammatory responses, but the connection between the compound and a specific disease was perhaps not as clear as in this case. There is also room to argue in that direction, as the reasons in Dasatinib FCA are clear on their face that “the ability to inhibit a biological target implicated in disease” is not necessarily the minimum required by the “scintilla”. Rather it is “ doubtlessly a useful discovery” and “certainly” sufficient to meet the standard. While not the final word on the subject, the guidepost established by the Court in Dasatinib FCA is both very helpful, and planted in the right spot.

No New Cases

No new patent / NOC / data protection cases were released last week.

Note that I generally only blog on substantive patent / pharma cases. If you want to keep abreast of all new Canadian decisions, including procedural decisions and copyright and trade-mark cases, I recommend subscribing to the Daily Intellectual Property News service from Alan Macek's IPPractice.

No New Cases

No new patent / NOC / data protection cases were released last week.

No New Cases

I'm back from vacation, but no new cases were released in the interim.

Blogging Break

I'll be taking a break from blogging for a couple of weeks. I'll resume after Labour Day, starting with any decisions that might have been released in the meantime.

Is the Tariff System for Costs Broken?

Dow Chemical Co v Nova Chemicals Corp 2017 FC 759 Fothergill J
            2,160,705 / film-grade polymers / ELITE SURPASS

In this costs decision related to the remedies phase of this litigation, Fothergill J awarded Dow, the successful plaintiff, legal fees in the amount of $1,200,000.00. This was about 15% of Dow’s claimed costs, though Fothergill J found that the legal fees claimed by Dow were not adequately supported by evidence, and did not on their face appear reasonable [20]. Of more interest than the size of the award is that Fothergill J used his discretion to assess the award at three times the maximum specified under the Tariff (approximately $400,000.00) [22]. As discussed here, O’Keefe J also departed from the Tariff in awarding costs in the liability phase, 2016 FC 91 saying the Tariff award would be “totally inadequate” [26]. This suggests that the Tariff system is so outdated as to be effectively broken, at least for major patent litigation. The Federal Court has long since recognized this, with a Sub-Committee having been struck almost three years ago, in November of 2014, and a discussion paper issued in October of 2015. Perhaps I’ve missed it, but I haven’t seen anything more from that Sub-Committee since then.

What is the Correct Date for Assessing Sufficiency?

Idenix Pharmaceuticals, Inc v Gilead Pharmasset LLC 2017 FCA 161 Rennie JA: Pelletier, Near JJA aff’g 2015 FC 1156 Annis J
            2,490,191 2,527,657 / sofosbuvir / SOVALDI
            FC Overview FC Sufficiency FC Inventorship FC ToC

At trial Annis J held the Idenix’ ‘191 patent invalid for lack of utility and insufficiency: see my posts on sufficiency and utility. Idenix appealed on both issues. The FCA affirmed on sufficiency [50], and consequently it was unnecessary to address utility [52]. As the Court noted, Idenix’s appeal “raises no new issues of principle or novel application of established principles to the facts” [3]. Idenix made the usual valiant attempt to unearth errors of law, but the FCA held either that the putative errors where not errors when read in context, or that they were inconsequential, or that were really an attempt to reargue the facts.

One point that caught my eye is the Court’s statement that sufficiency of disclosure is determined “as of the date of filing” [46], citing Viagra 2010 FCA 242, [79] (rev’g on other grounds 2012 SCC 60]. However, in Zoledronate 2013 FC 283, Hughes J, after consideration of the case law, held that the appropriate date for consideration of sufficiency was the date of publication, not the filing date [179]-[188] (aff’d 2013 FCA 244 without discussing this point). Kelen J 2009 FC 638, [108] in his Viagra decision held the same, relying on an earlier decision by Hughes J 2007 FC 596, [140-41] – though just as noted, on the same case on appeal, Viagra 2010 FCA 242, [79], the FCA held the correct date was the date of filing. (In that paragraph the FCA stated that some of Kelen J’s comments were “misguided” but it is not clear to me that the FCA was referring to his comments regarding the date for assessing sufficiency.) Hughes J’s logic in Zoledronate was that the purpose of the sufficiency requirement is to ensure that the invention is given to the public, and since it is directed at the public there is no point in considering a date before the public can actually see it. On the same logic, the SCC has held that the date for construing the claims, and more generally “the language of a patent” is is the date of publication: Free World 2000 SCC 66, [52]; Whirlpool Corp v Camco Inc 2000 SCC 67 [56]. It seems a bit odd to say that the words are construed at the date of publication, yet sufficiency is assessed at the same of filing, in light of how the words would be understood at the publication date, but it is perhaps not logically inconsistent, though there is the unifying point of principle that both are addressed to the public. I suppose the argument for using the filing date (which was not articulated in either FCA decision), is that the disclosure is not just addressed to the public, but also to the patent examiner; an application for a patent which does not sufficiently describe how to make the invention can be refused for that reason, and it seems wrong to hold the patent valid after it is granted, just because it slipped through the cracks on examination. That is particularly because if the patent was insufficient as of the date of filing, but sufficient at the date of publication, that will usually be due to the efforts of some party other than that patentee that has become common general knowledge.

I’m not really sure what to make of this. The decision of Hughes J in Zoledronate is the most fully considered decision on the issue, and the FCA holdings were stated without analysis of the case law. Moreover, in this case, nothing turned on the difference between the filing date and the publication date, and I believe the same was true in Viagra. But of course, the FCA is the higher authority. It’s probably a rare case in which the outcome turns on whether sufficiency is assessed at the claim date or the publication date, but it would be nice to have an FCA decision on this point which directly discusses Hughes J’s Zoledronate analysis, and so resolves the issue one way or the other.

Inducement to Induce Infringement

Elbit Systems Electro-optics Elop Ltd v Selex ES Ltd aff’d 2016 FC 1129 Martineau J aff’g 2016 FC 1000 Tabib J

2,527,754 / Fiber Laser Based Jamming System / MUSIC / MIYSIS

I don’t usually blog about motions to strike, since the bar for striking is so stringent that a refusal to strike is not very informative, but this one is worth a brief note as it raises a novel issue of potentially wide applicability: is inducement to induce infringement a permissible cause of action?

A subcontractor provided indemnity against patent infringement to a prime contractor who, in turn, provided an indemnity to the ultimate customer [24], [30]. The allegation is that but for the agreement to indemnify, the prime contractor would not have selected the subcontractor to provide the allegedly infringing product. If the prime contractor is inducing infringement, can the subcontractor be liable for inducing the prime contractor to induce infringement? The patentee is also suing the subcontractor, who is a direct competitor, for direct infringement. Both the prime contractor and the customer are alleged to directly infringe, but neither is named as a defendant, presumably because the patentee does not want to bring an action against a potential customer.

Tabib J refused to strike the relevant passages of the statement of claim on the basis that the defendant had not satisfy the court that a cause of action in “indirect inducement” did not have the slightest chance of success, and Martineau J affirmed [30].

Quantum, Proportionality, and Uncertainty in Punitive Damages

Airbus Helicopters SAS v Bell Helicopter Textron Canada Limitée 2017 FC 170 Martineau J [FC Damages]
            2,207,787 / helicopter landing gear

As I noted in my post on Airbus’ entitlement to punitive damages, most jurisdictions do not award enhanced or punitive damages for patent infringement, or for civil actions more generally, because, as explained by Lord Reid in Cassell & Co Ltd v Broome [1972] UKHL 3 [1972] 1 All ER 801, it is undesirable to expose a party to a quasi-criminal sanction when both entitlement and quantum are so vaguely defined. With respect to quantum, he noted “[t]here is no limit to the punishment except that it must not be unreasonable.”

In US patent law uncertainty regarding quantum is addressed by limiting the total award to a maximum of three times compensatory damages assessed: 35 USC § 284. This means that the punitive damages are limited to twice the compensatory damages. So, in this case, the award of $1,500,000, comprised of $500,000 in compensatory damages and $1,000,000 in punitive damages, would be the maximum allowable under US law.

In Canadian law there is no such hard limit, either with respect to patents or more generally, though this may be because punitive awards have tended to be more modest and less of often granted. In Whiten v. Pilot Insurance Co 2002 SCC 18, [70], the SCC recognized the problem of uncertainty relating to quantum:

[70 [T]he incantation of the time-honoured pejoratives (“high-handed”, “oppressive”, “vindictive”, etc.) provides insufficient guidance (or discipline) to the judge or jury setting the amount. . . . A more principled and less exhortatory approach is desirable.

See similarly [39]. Thus the purpose of the principled approach is to provide guidance in assessing quantum. Does it?

Record Quantum

The Dow Chemical Co v NOVA Chemicals Corp 2017 FC 637 Fothergill J
            2,160,705 / film-grade polymers / ELITE, SURPASS

In the liability phase of this action, Dow v NOVA (No 1) 2014 FC 844 aff’d 2016 FCA 216, O'Keefe J held Dow’s 705 patent related to advanced film-grade “mLLDPE” polymers, to be valid and infringed by Nova. He also held that Dow was entitled to damages under s 55(2) of the Patent Act for pre-grant ‘infringement’, and that Dow was entitled to elect between damages and an accounting for post-grant infringement. Dow subsequently elected an accounting [107]. Fothergill J’s subsequent decision in the remedies phase, Dow v Nova (No 2) 2017 FC 350, addressed various issues to allow the parties’ accountants to calculate the actual sums owed by Nova to Dow [6]. (See my posts on reasonable compensation, springboard profits, fixed costs, and interest.) The parties were then able to largely agree on the quantum, but they returned to the court for further direction on three issues, which were addressed in this decision.

The first issue was whether Nova was entitled to deduct capital depreciation expenses for the division used to make the infringing product [5]. Fothergill J held it was not so entitled, as a matter of interpreting his own prior decision, without adding any substantive analysis [8].

Second, in Dow v Nova (No 2) [165], Fothergill J had allowed Nova to deduct a “proportional amount” of fixed costs, which he considered properly attributed to the production and sale of the infringing products. But what is “proportional”? Dow proposed an allocation based on relative production volumes [10], which Forthergill J accepted, noting that while Nova’s alternative proposals relied on “potentially valid distinctions,” Nova had not adduced evidence to support them [13]. As I suggested in my previous post on fixed costs, the theoretically ideal method is probably to allow full deduction of opportunity costs. Allowing a deduction of a portion of fixed costs is a half measure between that and refusing deduction entirely. Even though not ideal, that approach may be justified on the basis of practical administrability as compared with a full opportunity cost approach. If it is correct to say that deduction of proportional fixed costs is justified on that pragmatic basis, it is not surprising that there is no single theoretically appropriate basis for the apportionment.

The third issue is the timing of the conversion of certain capital expenditures from Canadian dollars to U.S. dollars for the purpose of calculating annual capital depreciation [15]. Nova sold most of its infringing product in the US and retained its profits in US dollars (Dow v Nova (No 2), [189]), and Fothergill J held the profits to be disgorged should be calculated in US dollars and converted to Canadian dollars as of the date of judgement (Dow v Nova (No 2), [176]). Nova was entitled to deduct some part of the depreciation on the plant that made the infringing product, but that plant was built in Canada and paid for in Canadian dollars. The question was how the depreciated amount should be converted to US dollars to be deducted from the annual profit, which was calculated in US dollars. Dow’s position was that the appropriate date for conversion was when the expenditure was incurred, so that the full capital amount would be converted into US dollars at the exchange rate for that year, then annual depreciation would be calculated in US dollars [16]. Nova’s position was that the annual depreciation should be calculated in Canadian dollars, and then converted to US dollars on an annual basis, using the exchange rate for the year of the deduction, on the view that the point of depreciation is to recognize that capital-related expenses are multi-year expenses [16], [18]. The issue was important because the exchange rate had fluctuated substantially over the relevant period, and “Dow complains that [Nova’s] approach results in a greater deduction for depreciation in USD than the costs Nova actually incurred in USD,” [17] I’m not sure I’m entirely persuaded by Nova’s point, since no one was taking issue with the depreciation itself, and even if it is true that capital-expenses are multi-year expenses, it does seem strange that the amount of the depreciation should be greater than the actual cost incurred. On the other hand, the conversion rate when the plant was built was unusually low [20], and Forthergill J was concerned that using it as the single point of conversion “would be punitive,” and he therefore adopted Nova’s approach, which was reasonable and supported by the evidence.

In the result, Fothergill J awarded Dow just under $645 million including pre-judgment interest to April 7, 2017, the date of judgment in Dow v Nova (No 2), plus pre-judgment interest to the date of this judgment. This seems to be a record award in reported Canadian patent cases.

Promise Doctrine Abolished

AstraZeneca Canada Inc. v. Apotex Inc. 2017 SCC 36 rev’g 2015 FCA 158 Dawson JA; Ryer, Webb JJA rev’g 2014 FC 638 Rennie J
            2,139,653 / esomeprazole / NEXIUM

In a unanimous decision, delivered by Rowe J, the SCC has abolished the Promise Doctrine in the Canadian law of utility [2], [24], [36], [37], [51]. I might almost leave it at that. Under law developed in the Federal Courts, Canadian utility doctrine had two branches: the long-established and uncontroversial “scintilla” branch, and the more recent and controversial Promise Doctrine. Because the scintilla branch was never in dispute, and the Promise Doctrine never served any purpose that wasn’t already better served by some other provision of the Act, the Court was able to simply abolish the Promise Doctrine. The result is surgical. The Promise Doctrine was a tumour, but one which had not metastasized. AstraZeneca has cut out the tumour, while leaving the body of utility law unchanged.

The Court’s decision was not based on a narrow technical analysis of the details of the Act or precedent. Rather, it was decided on the basis of fundamental principles. Rowe J called the Promise Doctrine “excessively onerous” [37] and “punitive” [51], and called it “incongruent with both the words and the scheme of the Patent Act” [36], and “antagonistic to the bargain on which patent law is based,” as well as noting that it “undermines a key part of the scheme of the Act” [51].

What is the Promise Doctrine?

Under the law developed by the Federal Courts, the utility requirement had two branches, as summarized in Lilly v Novopharm / Olanzapine (No 1) 2010 FCA 197, [76] (quoted by the SCC at [29], emphasis added by SCC):

Where the specification does not promise a specific result, no particular level of utility is required; a “mere scintilla” of utility will suffice. However, where the specification sets out an explicit “promise”, utility will be measured against that promise:

The first branch represents the traditional utility requirement, which is now generally described as requiring a “scintilla” of utility. The second branch, emphasized by the SCC, is the Promise Doctrine.

A key feature of the Promise Doctrine is that the promise can be found anywhere in the specification [30]. This means that the promise of the patent may be, and in practice normally is, found in the disclosure, as opposed to the claims. Further, if there are multiple “promises” found in the disclosure, the invention must satisfy all of them, or the entire patent will be held invalid [31].

The Promise Doctrine can serve to elevate the requisite utility, often very substantially, above the standard that is required by the Act, as the FCA explained in Plavix 2013 FCA 186, [54]:

An inventor whose invention is described in a patent which would otherwise be valid can nonetheless promise more for his invention than required by the Act so as to render his patent invalid.

In this case, for example, the ‘653 patent claimed esomeprazole, which was a proton pump inhibitor (“PPI”) used for treating gastric disorders [3]. It was uncontested that esomeprazole was indeed useful as a PPI [9], and that use as a PPI was sufficient utility to support a patent [62]. The patent was nonetheless held invalid in the decisions on appeal, on the basis that the disclosure also promised that esomeprazole would be a better drug than existing drugs, and would work better for a wider range of patients [9]. That is more utility than is required for patentability, as it is well established that a new invention need not be better than existing product, if it “affords the public a useful choice”: Consolboard [1981] 1 SCR 504, 525. The only reason to hold the ‘653 patent to this higher standard was a putative “promise” found in the disclosure [63].

Promise Doctrine conflates disclosure and definition

The fundamental problem with the Promise Doctrine is that it conflates disclosure of the invention and definition of the invention [38]. These functions are distinct, and governed by different sections of the Act. The requirement that the patentee disclose the invention to the public is set out in s 27(3), while the requirement that the subject-matter of the invention “be useful” is set out in s 2 [40]. The specification consists of two parts. The first is a narrative portion normally called the “disclosure” (though it is now defined as the “description,”), that, as the name suggests, is primarily responsible for disclosing the invention. The specification ends with “a claim or claims defining distinctly and in explicit terms the subject-matter of the invention for which an exclusive privilege or property is claimed” (s 27(4)). Because the claims define the scope of the monopoly, it follows that validity must also be assessed against the claims: Whirlpool 2000 SCC 67 [49(b)]. Thus the description discloses the invention, the claims define the invention, and it is the invention as claimed, not whatever might have been disclosed, that must satisfy the requirements for receiving the protection of a patent, including the requirement for utility set out in s 2 [40]. Because the Promise Doctrine requires that an invention must satisfy the standard for utility set out in the disclosure, not just the requirement set out in s 2, the Promise Doctrine “effectively imports s. 27(3) into s. 2,” [44] thereby conflating those provisions [38].

Promise Doctrine wrongly invalidates patents which satisfy the requirements of the Act

This is not a technical point. Under the Promise Doctrine, an inventor can be denied a patent, or have a granted patent held invalid, even though the patent satisfies all of the statutory criteria. This happened regularly under the Promise Doctrine; this case is but one example. (See my article, The False Doctrine of False Promise, (2013), 29 CIPR 3, 33 (available here in good draft form) for a summary of the cases up to 2012.) This result is “incongruent with . . . the scheme of the Patent Act” [36], because in Canadian law, a patent is a statutory right; an inventor is entitled to a patent if the invention satisfies the statutory requirements for patentability: Harvard Mouse 2002 SCC 76, [11], [119]; Monsanto [1979] 2 SCR 1108, 119-20. The judge-made Promise Doctrine was contrary to the scheme of the Act because it could result in the inventor being denied a patent to which it was entitled by statute.

Promise Doctrine discourages full disclosure

Thus the Promise Doctrine was inconsistent with the fundamental nature of the patent right under the Patent Act, with the result that inventors were denied patents to which they were entitled by statute. This is bad enough, but the Promise Doctrine also has more insidious effects, extending to patents which are not actually invalidated by the doctrine.

The description and the claims are not just separate parts of the specification, they are functionally separate: the claims define the scope of the monopoly, while the description tells the public how to use the invention. The requirement to both define and describe the invention had long been a part of patent law, but at one time, both functions were served by a single narrative. The difficulty, as was explained over a century ago by Fletcher Moulton LJ, is that “These two things—the delimitation of the invention, and full practical directions how to use it—are in their nature almost antagonistic” (British United Shoe Machinery (1908), 25 RPC 631, 650 (CA), cited at [51]). The patent bargain requires “fulsome disclosure” of how to make and use the invention [51], but defining the invention requires narrow precision, to avoid encompassing subject matter which does not meet the statutory requirements for an invention. This tension was resolved by adding distinct claims to the end of the specification, which serve to define the invention. While now required by statute (s 27(4)), claims were originally implemented by patentees, for their own benefit, so that the requisite fulsome disclosure could be made without any risk that the disclosure would be taken to define the invention, with invalidity as a result. (For a more detailed discussion of the evolution of separate claims and their function, see False Doctrine of False Promise, 43-46.) Hence the rule that “what is not claimed, is disclaimed” (Monsanto 2004 SCC 34, [123]; Whirlpool 2000 SCC 67 [42]). This rule benefits the public, who thereby know the scope of the exclusive rights, but it also benefits the patentee, by excluding statements made in the disclosure alone from the definition of the invention. Under the Promise Doctrine, statements in the disclosure that are intended to describe the invention may be taken to define it, re-creating the very problem that gave rise to claims in the first place: “Thus, the Promise Doctrine undermines a key part of the scheme of the Act” [51].

By undermining the key distinction between disclosure and definition, the Promise Doctrine undermines the disclosure function: “To invalidate a patent solely on the basis of an unintentional overstatement of even a single use will discourage a patentee from disclosing fully, whereas such disclosure is to the advantage of the public” [51]. This chilling effect primarily affects those patents in which the disclosure is limited to avoid the risk of falling afoul of the Promise Doctrine. The abolition of the Promise Doctrine means that patent applicants can make full disclosure without worrying that statements in the disclosure will raise the bar on utility. And, as a practical matter, it means that patent applicants who have drafted their application primarily with the requirements of European or US law in mind will not be taken by surprise on this point when their patent is litigated in Canada.

The Court also held that the Promise Doctrine runs counter to the Act “by requiring that where multiple promised uses are expressed, they all must be satisfied for the patent to meet the utility requirement” [47]; and see also [37]. I see this as largely a corollary of the first point. So, the Court noted that this is potentially unfair because is risks invalidity if all of the multiple promises are not satisfied [50]. But given that a patent must have a scintilla of utility under the first branch to be patentable, even a single elevated promise effectively holds the invention to multiple utilities, namely the “scintilla” which is always required, and the elevated promise. Multiple promises are an exaggerated form of the same problem.

Consolboard is not authority for the Promise Doctrine

If, as the Supreme Court explained, the Promise Doctrine is fundamentally incompatible with the scheme of the Patent Act, how did it emerge in Canadian law in the first place? The answer is that it found its “new home in the Federal Courts’ jurisprudence” [35] on the basis of a narrowly technical misinterpretation of the case law.

The primary Canadian case cited as authority for the promise doctrine is the following passage from the SCC decision in Consolboard [1981] 1 SCR 504, 525 (my emphasis):

. . . the Federal Court of Appeal erred also in holding that s. 36(1) [now s. 27(3) and (4)] requires distinct indication of the real utility of the invention in question. There is a helpful discussion in Halsbury’s Laws of England (3rd ed.), vol. 29, at p. 59, on the meaning of “not useful” in patent law. It means “that the invention will not work, either in the sense that it will not operate at all or, more broadly, that it will not do what the specification promises that it will do”. There is no suggestion here that the invention will not give the result promised. . . .

This passage, and the underlined phrase in particular, is the central authority for the proposition that the promise doctrine was part of the law of Canada: see e.g. Olanzapine (No 1) quoted at [29]. Consolboard is the only pre-2005 Canadian case ever cited by the Federal Courts as authority for the promise doctrine: see False Doctrine of False Promise at 23. (Of course, after 2005, the Federal Courts began citing their own post-2005 case law.)

The SCC in AstraZeneca firmly rejected the proposition that Consolboard is authority for the promise doctrine:

[43] While the above passage [from Consolboard] uses the word “promise” it does not refer to, nor does it embody, the Promise Doctrine.

The problem is that the courts citing Consolboard as authority for the Promise Doctrine seized on the word promise, while ignoring the context. As the Supreme Court emphasized in Consolboard itself, the central point of the decision was that “the Federal Court of Appeal has confused the requirement of s. 2 of the Patent Act defining an invention as new and ‘useful’, with the requirement of s. 36(1) [now s 27(3)] of the Patent Act that the specification disclose the ‘use’ to which the inventor conceived the invention could be put” (527). Indeed, the reason the Court in Consolboard quoted the passage in question was to emphasize this distinction (526). A case that stands for the distinct nature of the disclosure and definition requirements cannot possibly be good authority for a doctrine which runs them together. As Rowe J noted in AstraZeneca:

[38] [T]he Promise Doctrine runs counter to the scheme of the Act by conflating ss. 2 and 27(3) — the very confusion this Court sought to clarify in Consolboard.”

The other pre-2005 authority cited by the Canadian courts was old English case law: see False Doctrine of False Promise at 23. However, that English case law is clearly not good authority for the doctrine in Canada, because the grant of a patent had a fundamentally different foundation than in Canadian law. When the English version of the Promise Doctrine developed, the grant of a patent was not a matter of right, but rather a discretionary exercise of the Crown prerogative, and the basis of the promise doctrine in English law was the “unwillingness of the courts to second-guess the Crown in the exercise of its discretion,” ([35], quoting False Promise of False Doctrine at 17.) As noted above, the grant of a patent is Canada is not discretionary, and it never has been: the Canadian Patent Act was originally based on the U.S. Patent Act, not on English law, and the non-discretionary nature of the grant of a patent in Canadian law follows the U.S. model. Thus the justification for the Promise Doctrine in English law is inapplicable in Canada. Indeed, citing the English case law as authority was especially misplaced because the Promise Doctrine is now “extinct” in English law [35]. It disappeared with the Patents Act, 1977, which formally made the grant of a patent a statutory right, as it always had been in Canada: see Siebrasse, Form and Function in the Law of Utility, (2015) 30 CIPR 109, 126-32 (available here in good draft form). So, Canadian courts were citing English cases as authority for a doctrine which no longer existed in English law. Moreover, the reason it no longer existed is that the scheme of the UK Act had changed to become more similar to that of the Canadian Act. In effect, it is backwards to cite the old English cases as authority for the Promise Doctrine in Canada; the real principle emerging from the English cases is that the Promise Doctrine is “incongruent” with a patent system, like the Canadian system and the modern English system, in which the grant of a patent is a statutory right.

Promise Doctrine serves no purpose not better served by other doctrines

The SCC noted that supporters of the Promise Doctrine assert that it is important to ensure patentees do not “overpromise” in their patent applications [45]. The Court rejected this argument in a brief paragraph. While overpromising is a “mischief” [45], this mischief is treated by the Act in multiple ways [46]. A patent that claims more than it discloses will be invalid for insufficiency under s 27(3), and if a statement in the specification is “willfully made for the purpose of misleading,” the patent will be void under s 53 [46].

I have seen some suggestions that this somehow opens the door to a more aggressive use of the disclosure requirement. In my view, that suggestion is entirely without foundation. This is not a call to arms. The SCC called overpromising a “mischief” not a scourge, and it gave a few examples – “for instance” – of the way the law already “treats” that mischief [46]. The Court gave not the slightest suggestion that these existing treatments were inadequate. Nor, really, did it say the contrary. The Court’s explicit point is that the Promise Doctrine is not the appropriate mechanism for addressing this issue, and other mechanisms exist. To read anything more into the brief discussion would be an error – indeed, the same type of error as was made in building the Promise Doctrine out of a single acontextual phrase from Consolboard.

Correct approach to utility

The SCC’s legal analysis concluded by setting out the correct approach to utility. There is nothing surprising in this discussion, given that the Promise Doctrine has been rejected.

The Court reaffirmed that a “scintilla” of utility is all that is required. While that term has been commonly used in the Federal Court for the last decade, I believe this is the first time the SCC has used the term “scintilla” to describe the standard for utility. It is nonetheless consistent with the older jurisprudence holding that “very little will do”: see e.g. Wandscheer [1948] SCR 1, 24. The Court also noted that “A single use related to the nature of the subject-matter is sufficient.”

The Court also affirmed the holding from Consolboard that “a patentee is not required to disclose the utility of the invention to fulfill the requirements of s. 2” [58]. However, in my view, when the utility would not be evident to a skilled person on the basis of their common general knowledge, or from the nature of the claimed invention itself, as may be the case for a new chemical compound, it may be necessary to disclosure the utility in the specification: see Siebrasse, Must the Factual Basis for Sound Prediction Be Disclosed in the Patent?, (2012) 28 CIPR 39, 56-59. Arguably this is required not as a matter of s 2, which requires only that the invention be useful, but rather to satisfy the disclosure requirement that the inventor describe the invention so that persons skilled in the art can “use the invention as successfully as the inventor could himself” (Consolboard, 526, quoting R. v. American Optical Company). In any event, what is more important for present purposes is that AstraZeneca does not change anything on this point, one way or the other. The question of whether utility has to be disclosed was not at issue in this case, and the Court is simply reaffirming the point already made in Consolboard.

The Court stated that “utility must be established by either demonstration or sound prediction as of the filing date,” citing AZT [55] (my emphasis). This is significant in affirming the filing date, not the priority date, as the appropriate date for assessing utility. AZT itself was ambiguous, or rather inconsistent on this point, referring to both the filing date and the priority date as being the appropriate date. (There was no debate between the parties in AZT on this point, which no doubt accounts for the confusion). Subsequent Federal Court decisions have settled on the filing date: Ramipril 2005 FC 1283, [91]-[96] aff’d 2006 FCA 64, [30]; Quinapril 2007 FCA 209, [153]. Here, the SCC is apparently accepting the filing date as the correct date. The point was not argued in this case – both parties accepted the filing date as the correct date for establishing utility – so the authority is somewhat weaker than if the point had been in dispute, but given the consistent FCA holdings and this statement by the SCC, the point seems settled.

The Court noted that the purpose of the utility requirement is to “avoid granting patents prematurely, and thereby limiting potentially useful research and development by others” [56]. This is a helpfully clear statement, which is consistent with prior caselaw: see e.g. AZT, [56], Wandscheer v Sicard Ltd, [1948] SCR 1, 5, 10. It is also consistent with the position in the US, Europe, and the UK: see Brenner v Manson, 383 US 519 (1966); T 0870/04, BDP1 Phosphatase / Max-Plank, [21]-[22]; T 0898/05 Hematopoietic cytokine receptor / Zymogenetics, [7]; HGS v Lilly, [2011] UKSC 51, [ 91], [102], [107]. This raises the question of how early is too early? The Court went on to say that the utility requirement “is to be interpreted in line with its purpose — to prevent the patenting of fanciful, speculative or inoperable inventions” [57]. This indicates that the line is drawn quite far upstream, as it is only “fanciful, speculative or inoperable inventions” which are excluded. This is consistent with the low “scintilla” standard for utility. But again, we should not read too much into this statement. The question at issue in AstraZeneca was the Promise Doctrine, not the question of how high the bar is for the scintilla standard. The Court’s general statement reaffirms the prior case law and provides helpful guidance in principle, but it does not address, nor does it purport to address, such details.

One general point made by Rowe J is that a use which is acceptable to satisfy the utility requirement must be related to the claimed subject-matter: “a proposed invention cannot be saved by an entirely unrelated use” [53]. That general statement is clarified by the example: “It is not sufficient for a patentee seeking a patent for a machine to assert it is useful as a paperweight” [53]. While I’m not sure I’ve seen that point made in quite this way before, the principle is entirely consistent with established Canadian law; no doubt the reason it has not arisen directly is that no patentee has had the temerity to argue, for example, that a new chemical compound with no known use should be considered useful as landfill. It corresponds to the requirement in US law for a “specific” utility – since any machine could be used as a paperweight, such a use would not be considered specific to the invention (unless, of course, its main purpose was as a paperweight): see Brenner v Manson, 383 US 519 (1966).

The Court also set out a two step analysis for approaching the utility requirement:

First, courts must identify the subject-matter of the invention as claimed in the patent. Second, courts must ask whether that subject-matter is useful — is it capable of a practical purpose (i.e. an actual result)?

The first point is that the utility inquiry is directed to the invention as claimed. In context, this is in contrast with assessing utility on the basis of the invention as disclosed in the specification. The second branch just repeats the utility requirement.

The result in this case

Rennie J at trial had accepted that esomeprazole was useful as PPI (on the basis of sound prediction) [62], and he had held the patent invalid solely on the basis of the failure to meet the elevated promise [62]. Consequently, with the Promise Doctrine disposed of, it was straightforward for the Court to hold the ‘653 patent valid [63], without any need for remand.

Conclusion

The most obvious consequence of AstraZeneca is that patents for good inventions will no longer be invalidated on the basis of a doctrine which “has no basis in the Act” [51]. Legal uncertainty will be substantially reduced, as the actual process of determining the promise of the patent was unpredictable and arbitrary, as I described in The False Doctrine of False Promise, at 35-40. It often happens that when a case reverses or clarifies a problematic doctrine, there is a subsequent period of uncertainty as the law settles into its new form. That should not be a concern in this instance. Because the Promise Doctrine was so idiosyncratic and out of step with the scheme of the Act, its excision leaves no gap to be filled. With the abolition of the Promise Doctrine, Canadian patent law is also brought more into line with our major trading partners. (For example, the Promise Doctrine is one reason the US put Canada on the Special 301 Report Watch List. The SCC decision entirely resolves that concern.)

This does not mean that all is well in Canadian utility law, or that no differences remain between our law and European or US law. Outside of Canada the phrase “promise doctrine” is sometimes used to refer to the entire constellation of quirks in Canadian utility law, but Promise Doctrine dealt with in AstraZeneca refers only to an elevated standard for utility imported from the disclosure. In particular, one point of considerable controversy is the enhanced disclosure requirement for utility based on sound prediction, which requires the factual basis and sound line of reasoning supporting the prediction to be set out in the patent itself (see generally my article, Must the Factual Basis for Sound Prediction Be Disclosed in the Patent?). That doctrine is unaffected by this decision. (The issue was raised at first instance, but only in obiter, and it was not addressed by the Supreme Court.) I have seen it suggested that AstraZeneca overturns that enhanced disclosure requirement, on the basis of the Court’s statement at [58], that a patentee is not required to disclosure utility. The argument, presumably, is that if the patent must disclose the factual basis for utility, the patent must necessarily disclose the utility, but since disclosure of utility is not required, it must follow that disclosure of the factual basis for utility is not required. While there is logic to this reasoning, in my view it is not plausible that the Court intended to eliminate a doctrine that is controversial yet well-established at the Federal Court level, purely by implication, without even referring to it directly. Perhaps there is an inconsistency between the enhanced disclosure requirement and the rule that the patentee is not required to disclose the utility of the invention, but if so, that inconsistency existed ever since the enhanced disclosure was developed, which was long after Consolboard. Put another way, if the enhanced disclosure is bad law, that is because it has always been bad law, and not because AstraZeneca has changed the relevant law. The Promise Doctrine arose as a result of an acontextual over-interpretation of a few words in Consolboard. That is a mistake which should not be repeated in interpreting AstraZeneca.