PM(NOC) Statutory Stay Damages Regime is a Complete Code

Apotex Inc v Eli Lilly Canada Inc 2021 ONSC 1588 Schabas J

            2,041,113 / olanzapine / ZYPREXA

Under the patent linkage system established by the PM(NOC) Regulations, a patent that is ultimately held to be invalid can keep competitors off the market for two years by operation of the statutory stay pursuant to s 7(1)(d). If the generic prevails, s 8 provides a remedy in the form of damages for the losses suffered from having been kept off the market by the statutory stay. There are limitations to the s 8 remedy which are now clearly established. First, if the generic prevails in the s 8 proceeding, s 8 entitles it only to its own damages, and not disgorgement of the patentee’s profits: 2011 FCA 358 (here); 2013 ONCA 555 (here). Second, if the generic is unsuccessful in the NOC proceeding, it cannot claim s 8 damages, even if the patent is subsequently held invalid in an infringement action: 2013 FCA 282 (here). A number of actions have been brought in provincial superior courts pleading a variety of causes of action (other than s 8), seeking a way around both of these limits. So far, none of these attempts to avoid the limits on s 8 recovery have been successful, though some causes of action have survived a motion to strike. Other causes of action have been struck, or class certification refused on substantive grounds: see here for a summary as of Aug 2016, and here and here for discussion of two decisions since then. (I think that is a complete list, though I won’t make any promises.)

In this case, Apotex sought to avoid the second limit on s 8 damages. Lilly had prevailed in NOC proceedings against Apotex ( 2007 FC 455 affd 2008 FCA 44), but 113 patent was subsequently declared invalid in separate proceedings (2011 FC 1288 affd 2012 FCA 232): [10]–[23]. (As Schabas J noted [23], the 113 patent was invalidated on the basis of the promise doctrine—see here—which was subsequently abolished by the SCC in AstraZeneca 2017 SCC 36, so it appears that the 113 patent would be valid under current law. Nothing turned on this point.) Apotex had attempted to claim s 8 damages in Federal Court proceedings, and it was that litigation, culminating in 2013 FCA 282, that established that s 8 recovery is not permitted in those circumstances.

In this action, Apotex sought to recover damages for having been kept off the market on the basis of three causes of action: breach of the Ontario Statute of Monopolies; s 7(a) of the Trademarks Act; and common law conspiracy in restraint of trade [34]–[36].

Schabas J dismissed the action on a motion for summary judgment, holding that none of these causes of action raised a genuine issue for trial. There were two preliminary points: whether the matter should be dealt with by way of summary judgment (yes), and whether the limitations period had run (no), which I won’t address, as the former is not patent specific and the latter is only relevant to the old NOC proceedings.

Schabas J’s holding rested primarily on two general points.

First, the PM(NOC) Regulations are a “complete code” which comprehensively sets out the remedies available to the parties. The underlying NOC proceedings in this case were standard NOC proceedings of the kind that would have been directly within the contemplation of the legislature. To provide additional remedies which are not found in the Regulations, in circumstances that undoubtedly were contemplated by the drafters, would upset policy choices and balancing of interests arrived at by Parliament: [110] and generally [97]–[110], [114]. As Schabas J pointed out, this reasoning has been accepted in all of the highest authorities to consider the question in this context.

In my view this logic is sound. Since I’m an academic, I have to quibble a tiny bit. Schabas J stated a few times, including at the very outset of his analysis, that “patent law is ‘wholly statutory’” [97], [110]. Yes, that is quoting the SCC in Sanofi 2008 SCC 61, which in turn quoted Lord Walker’s decision in Synthon [2005] UKHL 59, but Apotex nonetheless is right to say that this is an “overstatement” [98]. For example, all of the law of utility, including the doctrine of sound prediction and its enhanced disclosure requirement (if that exists), and the prohibition of post-filing evidence, rests on a single word in the Act; the obviousness requirement was not in the code at all until 1993, despite having been part of the law for a century before that. The Supreme Court in Sanofi used this phrase to indicate two somewhat narrower points: first, that there is no common law right to a patent; and second, that specific statutory provisions will always prevail over judicially developed doctrines. Both of these points are unassailable, and neither rests on the notion that the Act is a complete code in a strong sense. Lord Walker’s statement in Synthon [57] is particularly curious:

The law of patents is wholly statutory, and has a surprisingly long history. It has been wholly statutory since the Statute of Monopolies 1623, an important landmark in constitutional history because of its effect in curbing the royal prerogative.

As Lord Walker states explicitly, the Statute of Monopolies did nothing but curb the royal prerogative. Even after the Statute of Monopolies, the basis for the grant in English law rested on the exercise of the royal prerogative and not on the statute: in principle, this was true in the UK until the Patents Act 1977. The Crown could, and occasionally did, refuse to grant patents for general policy reasons, such as protection of tax revenue, which had not even the slenderest basis in the Statute of Monopolies. Thus the very basis of the grant was not statutory. Moreover, the Statute of Monopolies itself is and was at the time generally considered to be a reflection of the common law. Indeed, Dr Fox has gone so far as to say that the Statute of Monopolies “was not by any means enacted as a patent statute having as its primary purpose the establishment of a system whereby letters patent might be obtained for meritorious inventions. It was enacted as the result of a long and turbulent agitation against the abuse of monopolies of all kinds as well as other grievances which are set out in the preamble to the Act”: Fox, “Abuse of Monopoly”(1945) 23 Can Bar Rev 353, 371. Consequently, Lord Walker’s statement that patent law “has been wholly statutory since the Statute of Monopolies 1623" is a bit of a stretch.

With that said, Schabas J’s decision does not rest on the view that the entirety of patent law is statutory. It rests on the much narrower ground that the balance of interests struck in the PM(NOC) Regulations, and the remedies provisions in particular, are a complete code for addressing the harm done by the statutory stay. That is the main point made by the authorities quoted and relied on by Schabas J. This narrower point is entirely compelling. The PM(NOC) Regulations were high profile enactments which were heavily lobbied. They reflect a political compromise between the interests of the generics and the innovators, and more broadly “a compromise between the interests of the public in encouraging research and development of new patentable drugs and in encouraging generics to market drugs at lower prices” [106]. I won’t multiply references on this point, as it is made in all of the decisions quoted by Schabas J, noting eg the “highly regulated” nature of this field [109].

I’ll add one point that wasn’t mentioned in the decision. It is widely recognized that the statutory stay "is tantamount to an interlocutory injunction": Bayer AG v Apotex (1993) 51 CPR(3d) 329 [13]. While the 1993 RIAS is not explicit, it appears that the NOC proceedings under the original regulations were broadly modeled on an interlocutory injunction application (in contrast to the deemed infringement approach under the US Hatch-Waxman system and under the new NOC proceedings). So, in addition to the stay, the proceeding was summary in nature, and the fact that the outcome had no in rem effect is also similar to an interlocutory injunction application. A defendant may also suffer loss from being enjoined by an interlocutory injunction if it turns out that the injunction was ‘wrongly’ (with the benefit of hindsight) granted, and it is very well-established that the party subject to an interlocutory injunction has no common law right to damages for its loss. This is because the loss is due to an order of the court, not the action of the plaintiff. The defendant will only be compensated if an undertaking is given, as is normally required. Section 8 damages play the role of damages on the undertaking; and to the extent that the proceedings were indeed modeled on an interlocutory injunction, the parallel implies that the background assumption is that there would be no recovery other than that provided for by s 8, just as there is no recovery for the harm caused by an interlocutory injunction apart from that provided for by the undertaking.

The second general point made by Schabas J is that Lilly cannot be made liable for exercising its statutory rights [113]–[118]. The direct cause of the loss to Apotex is the statutory stay. The stay is indirectly caused by Lilly listing the 113 patent, and then responding to Apotex’s NOA. While Lilly was not required to list the 113 patent, it was entitled to do so: “such acts ‘were the opposite of unauthorized’ and were acts that Lilly ‘was at liberty to commit’” [118], quoting Harris 2010 ONCA 872 [50] affg 2010 ONSC 2326 [94]. This is true “even if Lilly was motivated by an improper motive” (on which there is no evidence) [111].

This reminds me broadly of the US Noerr–Pennington doctrine, which is to the effect that it is not actionable under the antitrust laws to petition government to enact anticompetitive legislation, even if the intent is to eliminate competition: Noerr 365 US 127 (1961); Pennington 381 US 657, 670 (1965). The parallels are not exact, as Noerr and Pennington themselves are about petitioning government in the antitrust context specifically. But there is a broad parallel in the idea that petitioning the government in respect of legislation cannot be actionable in a democracy (Noerr 137–38), and similary taking advantage of one’s rights under legislation should not be actionable. The lower federal courts in the US have consequently expanded that immunity far beyond the antitrust context to a wide variety of claims: see Gugliuzza, Patent Trolls and Preemption, (2015) 101 Va L Rev 1579, 1611–12. Moreover, the Court noted in Pennington 671 that “it is clear under Noerr that Phillips could not collect any damages under the Sherman Act for any injury which it suffered from the action of the Secretary of Labor. The conduct of the union and the operators did not violate the Act, the action taken to set a minimum wage for government purchases of coal was the act of a public official who is not claimed to be a co-conspirator, and the jury should have been instructed, as UMW requested, to exclude any damages which Phillips may have suffered as a result of the Secretary’s Walsh-Healey determinations.” That is, if the direct cause of the injury is the operation of the legislation, the injured party cannot claim damages against the legislator, and any improper motive in petitioning for the legislation is vitiated by the intervention of the legislator itself. The principle applies not only to petitioning a legislative body, but also petitioning a court for relief, unless the litigation is a “sham”: see Professional Real Estate Inventors v Columbia Pictures 508 US 49, 56–58 (1993). More broadly, in the United States under Noerr-Pennington and related doctrines, a patentee is generally immune from civil liability for harm arising from assertions of infringement so long as the assertions are not (1) objectively baseless and (2) made with subjective ill intent: see Gugliuzza, Patent Trolls and Preemption, 1616–28. While there is room for debate about whether the precise contours of the US rule are sound as applied in the patent context (see Gugliuzza, arguing that threats made in bad faith should not be immune from liability), that is not relevant here as there is no question of objective baselessness or ill intent on the facts at hand. (Indeed, as noted above, the patent at issue would probably be valid under existing law.)

I’m not sure how much this parallel adds, but in any event, I do find the basic point made by Schabas J and Harris—that it cannot be actionable to take advantage of one’s statutory rights—to be entirely convincing.

It seems to me that these points are independent, and either would be sufficient to dispose of Apotex’s claim. The policy balance struck by the legislature happens to be triggered by the pharma innovator listing its patent; but even if there were some other trigger, the policy balance still shouldn’t be upset. Conversely, if listing is authorized under the Act, it shouldn’t give rise to liability, even if there weren’t a comprehensive balance backing it up. But that’s an issue for another day, as here the two principles did work in tandem.

So at this point there is a strong argument for dismissing Apotex’s action—and I haven’t even discussed the actual causes of action themselves, except in passing at the outset. Schabas J’s decision proceeded similarly: he apparently held that the claims were not tenable even before considering the specific causes of action [122], though he did go on to consider the specific causes of action, and concluded that none of them could be supported on the evidence or the law. While this approach is a bit unorthodox, it strikes me as sound. At the very least, the general principles establish that the claims should be dismissed unless examination of the specific causes of action shows some good reason to the contrary.

The first cause of action was for breach of s 4 of the Ontario Statute of Monopolies, which provides (in part),

If any person shall be . . . grieved . . . by [any] letters patents . . . tending as aforesaid, . . . the same person shall have his remedy . . . against him by whom he shall be so . . . grieved . . . and every such person . . . shall recover three times so much as the damages which he sustained by . . . occasion of being so . . . grieved

This is essentially identical to s 4 of the original English Statute of Monopolies (barring some transitional language in the original statute). As I understand it, there was some controversy as to whether various Imperial statutes affecting property and civil rights became part of the law of Canada ex proprio vigore. To resolve the uncertainty, “An Act respecting the Imperial Statutes relating to property and civil rights” SO 1902 c 13, was enacted, confirming that certain listed Acts were part of the law of Ontario. The Ontario Statute of Monopolies was made part of the RSO 1897 under the authority of that Act: see Fox, “Abuse of Monopoly” (1945) 23 Can Bar Rev 353, 363–64. The Ontario Statute of Monopolies has never been judicially considered until this recent spate of proceedings. Section 4 of the original English statute appears to have been judicially considered exactly once in the 400 years since it was enacted, in Peck v Hindes (1898) 15 RPC 113 (QB) (see 117 indicating that it was the first time the provision had been considered).

There are serious questions as to the validity of the Ontario Statute of Monopolies, which Schabas J found it unnecessary to address in light of his conclusion that the provision, even if validly enacted, did not support Apotex’s claim in the circumstances [127]–[130]. I’ll note that Dr Fox was firmly of the view that it was intra vires: see “Abuse of Monopoly” at 370–72

Apotex’s argument, as I understand it, is that Apotex was “grieved” by the 113 patent, by virtue of having been kept off the market. Section 6 of the Statute of Monopolies (s 5 of the Ontario Statute) provides that s 4 does not apply to any letters patents granted to the true and first inventor of for new manufactures; but this exception, so the argument goes, should be interpreted as only applying to valid letters patent, so that patents that granted but which are ultimately held to be invalid are subject to s 4. This is a purely textual interpretation of s 4, and it must be said that a purely textual parsing of an obscure 400 year old statute is a very slender ground for disrupting the legislative balance struck by the NOC Regulations.

The historical intent of s 4 is obscure. In Peck v Hindes, counsel argued at 124–25 that s 4 was aimed at extrajudicial enforcement of patent rights; that was not accepted (or rejected) by the court, and even if correct, that theory of s 4 would not support a cause of action on the present facts. Mathew J in Peck v Hindes concluded at 127 that s 4 did not apply at all to letters patent for invention. The famous s 6 of the original Act (s 5 of the Ontario Statute, as s 5 of the original Statute was transitional), the foundation of the law of patents, provides an exception for letters patent for invention in the following terms: “any declaration before mentioned” —including s 4—”shall not extend to any letters patents. . . made. . .of the sole working or making of any manner of new manufactures.” The Court in Peck v Hindes observed that “Section 6 does not say anything about valid Letters Patent. It says, Letters Patent for new inventions” (127). Schabas J agreed with this holding [135].

The only scholarship I have been able to find that considers the effect of s 4 is series of articles by Dr Fox, primarily “Abuse of Monopoly”(1945) 23 Can Bar Rev 353; and see also “Patents in Relation to Monopoly” (1946) 12 Can J Econ Pol Sci 328; Fox, “Patents in Relation to Monopoly: A Rejoinder” (1947) 13 Can J Econ Pol Sci 68-80. Fox argues that s 4 was intended to address unjustified threats of legal proceedings, similar in spirit to the UK Intellectual Property (Unjustified Threats) Act 2017, and which are normally dealt with in current Canadian law under s 7(a) of the Trademarks Act: see here and here. While Dr Fox’s theory was not considered by Schabas J, Fox’s conclusions are entirely consistent with his reasons. Fox agreed with the holding in Peck v Hindes, saying that it is authority for the proposition that “an action for infringement of a patent, whether valid or invalid, does not give rise to the action contemplated by sec. 4” (370, my emphasis); and similarly, on the authority of Peck v Hindes, "a person could not be penalized for bringing action in a court of law to support his letters patent even though he were unsuccessful” (369). Fox goes on to argue that “nevertheless he could be and would be penalized if he endeavoured to use his patent as a club held in terrorem over the heads of his competitors, and by so doing grieved, disturbed or disquieted those persons, by methods other than bringing the matter for hearing and determination before a court of law” (369). Whether Fox is right as to the intent and effect of s 4 is a question that we need not resolve here; it is enough to say that on Fox’s view, s 4 would clearly not allow recovery for damages in the circumstances of this case.

It is also worth noting that Fox explicitly agreed with Schabas J’s second general point: “No person can be penalized, other than by costs, for bringing an action to support what he considers to be a legal right” (372). This observation illustrates that s 4 is not in tension with the general principles set out by Schabas J; on the contrary, it can, and should, be interpreted harmoniously with that basic principle.

The next cause of action raised by Apotex is based on s 7(a) of the Trademarks Act, which is breached when one makes “a false or misleading statement tending to discredit the business, goods or services of a competitor.” Apotex argued that the false statement by Lilly lay in listing the 113 patent on the Patent Register, which entailed a representation that the '113 Patent was valid and enforceable [139]. Schabas J noted that “there is nothing untrue or false in any material respect, let alone disparaging or discrediting of Apotex, in the Form IV. At the time it was completed, Lilly did have the 113 Patent for Olanzapine. That is essentially all that was stated” [141].

Apotex’s final cause of action is common law conspiracy, which alleged a conspiracy to fix monopolistic prices. This was premised entirely on Lilly applying for a patent and listing on the register, which Lilly was perfectly entitled to do. There was no suggestion of improper conduct, other than obtaining a patent that was ultimately held to be invalid; this cannot constitute a conspiracy.

Finally, I note that Schabas J did not hold that use of the litigation process can never give rise to civil liability, but rather that any such liability must be based on a “stand alone cause of action” or a claim “totally independent of the regulatory regime”: [115], adopting the language of Apotex v Eli Lilly 2015 ONCA 305 [53] and 2013 ONSC 5937 [8]. Schabas J did not elaborate, as there was no allegation capable of supporting any independent ground, but in my view, groundless threats of infringement, that were the concern of authors such as Dr Fox and Prof Gugliuzza, would certainly fall into that category. There is therefore not the slightest conflict between Schabas J’s holding and the FC decisions, mentioned above, dealing with unjustified threats under s 7(a) of the Trademarks Act.

PS — Various similar actions are also before the courts at various stages of case management. In a case management decision in one such matter, Apotex Inc v Pfizer Ireland Pharmaceuticals 2021 ONSC 1860 Myers J adjourned a trial set for September in light of Schabas J’s decision in this case, and instead ordered that the matter be decided by way of a motion for summary judgement.