Gillette Defence and Prior User Rights

Kobold Corporation v NCS Multistage Inc 2021 FC 742 Zinn J

            2,919,561

At one level this is a mundane procedural decision denying a motion to file expert reply evidence on the basis that the issue ought to have been anticipated [10]. But the effect is to hold that a Gillette defence cannot be based on prior use rights found in s 56 of the Act, with the consequence that the underlying motion for summary judgment was also derailed [19]. This is problematic in light of the “culture shift” in favour of simplified procedures endorsed by the SCC in Hryniak v Mauldin 2014 SCC 7 [2] that has been embraced by the Federal Courts in several recent decisions, most recently Canmar Foods 2021 FCA 7 affg 2019 FC 1233 (here). I suggest that a purposive analysis of the provision in question indicates that there may be a way to permit a motion for summary judgment to proceed in similar future cases.

After being sued for infringement of the 561 patent by Kobold, NCS brought a motion for summary judgment on the basis that the allegedly infringing procedure it currently uses is the same procedure that it had used prior to the claim date [5], [6]. NCS’s argument is similar to the so-called “Gillette defence,” established in Gillette Safety Razor (1913) 30 RPC 465 (HL), which permits the defendant to prove that the alleged infringement was practised in the prior art, so that either the alleged infringement does not fall within the claims of the plaintiff’s patent, or if it does, the patent is invalid for anticipation. (The Gillette defence may also turn on obviousness.) As Locke JA explained in Western Oilfield 2021 FCA 24 [77] the Gillette defence is not actually a distinct defence under the Act, but is rather “a shortcut around the often difficult and time-consuming process of construing the claims of a patent, and then determining whether those claims are valid and whether they have been infringed.” So, if the defendant can prove that it was only doing what it had always done publicly, which is purely a matter of fact, the claims asserted against it must either be invalid or not infringed. Crucially, this makes it unnecessary to construe the claims and assess anticipation; the whole point of the Gillette defence is that it avoids the cost of construing the claims and assessing validity in favour of a purely factual inquiry. The Gillette defence is often less helpful in practice because it is rare that a defendant will be willing to put all its eggs in that basket—but this suggests that the Gillette defence is a prime candidate for a summary trial, where determination of a single factual issue can determine the case.

That was NCS’s position in this case. NCS argued that the procedure it uses today is the same procedure it used prior to the claim date. But NCS did not rely on the Gillette defence as such. The allegedly infringing device was a bottom hole assembly used in oil and gas drilling and I suspect that NCS was concerned that it might not be able to establish that its prior use had made its device available to the public so as to be prior art for the purposes of anticipation under s 28.2. NCS instead sought to rely on the prior user rights granted by s 56, the effect of which is to allow a person to continue to do any act that it had done prior the claim date, even if its prior use had not been public.

More particularly, s 56(1) provides that “if—before the claim date of a claim in a patent—a person. . . committed an act that would otherwise constitute an infringement of the patent in respect of that claim. . . it is not an infringement of the patent. . . if the person commits the same act on or after that claim date.” The difficulty for NCS lay in the phrase “that would otherwise constitute an infringement of the patent.” In support of its motion for summary judgment based on s 56, NCS filed a fact affidavit of one its employees, evidently to the effect that NCS’s allegedly infringing bottom hole assembly was the same as NCS had used prior to the claim date [3]. In response, Kobold filed an expert affidavit going to infringement and (apparently) to claim construction [4]. NCS then brought a motion seeking to file expert reply evidence.

The Rules do not expressly allow for evidence in reply and such evidence is permitted only in special circumstances: Amgen v Apotex 2016 FCA 121. NCS argued the reply evidence should be permitted because it could not have anticipated that Kobold would file evidence on claim construction, because claim construction was irrelevant to the question of whether the bottom hole assembly it currently uses is the same as the assembly it used before the claim date [6]. Zinn J rejected this argument, saying that issues related to claims construction and whether NCS’s device was infringing ought to have been anticipated by NCS:

[10] These are two of the very issues at the heart of this litigation. Moreover, they are the very issues to be addressed when relying on section 56 of the Patent Act, unless a party admits that its tools and procedures do breach the relevant patent, but argues that they have been in use prior to the patent claim date.

The key point is that s 56 is available only in respect of “an act that would otherwise constitute an infringement.” As Zinn J put it, NCS’s position

ignores the express wording of section 56, which. . . speaks to a defendant committing an act before the claim date “that would otherwise constitute an infringement of the patent in respect of that claim.” This language brings directly into issue both claims construction and infringement. [11]

The effect of this holding is that a Gillette defence is available based on a prior public use, but not based on a prior secret use; while a defendant can avoid addressing claim construction and infringement when its prior use was public, it cannot avoid addressing those issues when its prior use was secret.

This also effectively means that a summary judgment is not available based on prior user rights. If the NCS’s device would not constitute an infringement, s 56 is not available, and unless NCS admits infringement, validity and infringement are therefore in issue. To get a summary trial focusing on the purely factual issue of prior use, NCS must admit infringement. It is of course a very risky proposition to put all its eggs in that basket.

This is somewhat counter-intuitive. In either case, the argument is that the defendant should be permitted to keep doing what it had always been doing, on the basis that the act must either be non-infringing, or permitted, either because the patent is invalid by anticipation, in the former case, or by prior user rights under s 56, in the latter case. Why should there be a functional difference between the two scenarios?

To answer that question we should look to the purpose of the key phrase. Why does s 56 apply only to an act “that would otherwise constitute an infringement of the patent”?

The long-standing provisions of the Act relating to prior user rights were replaced by entirely new provisions in 2018. It appears that the new provisions were modeled on the corresponding provisions of the UK Patents Act 1977, namely s 64. While there are some significant differences in the overall provision, this particular passage is very similar. It appears that the intent of this phrase in the UK Act is to ensure that prior uses which are non-infringing by virtue of statutory exceptions under s 60(5) of the UK Act, eg an experimental use, are not protected by the prior use provisions: see Terrell on Patents §14-203. It seems reasonable to suppose that the purpose of the parallel phrase in the Canadian Act is the same. (The main exceptions under the Canadian Act appear to be use authorized to respond to a public health emergency under s 19.4(7); regulatory use under s 55.2; and experimental use under s 55.3. There are also third party rights under s 55.11 which give rise to an exception under certain circumstances, but these have their own prior use provisions.)

That is, it appears that the intent of the key phrase is to ensure that the statutory exceptions to infringement do not give rise to prior user rights. An experimental use, for example, is a limited exception to infringement, which does not confer unlimited rights to make, sell and use the invention; consequently, an experimental use prior to the claim date should not give rise to a general right to make, sell and use the invention after the claim date. The defendant who was engaged in experimental use prior to the claim date is nonetheless entitled to continue engaging in the same limited experimental acts after the claim date, not by virtue of the prior user rights provisions, which are inapplicable, but by virtue of the experimental use exception itself. Extending the prior user rights acts to prior experimental use is therefore both unnecessary and undesirable.

If that is indeed the purpose of the key phrase, I suggest that the defendant need not admit infringement in order to take advantage of s 56; it should be enough for the defendant to admit that its acts do not fall within any statutory exception to infringement. On a motion for summary judgment, if NCS could then establish the factual point that the prior use was the same as the current use, while admitting that it was not subject to any exception—including eg that the use was not experimental—this would establish that NCS was entitled to the prior use defence without requiring it to admit infringement.

Bill C-86 and Prior User Rights

As I work my way through the patent provisions of Bill C-89, I’ll be writing some posts with my initial impressions. This post looks at the prior user rights provisions set out in ss 188, 194 of the Bill. (My last post looked at the amendments in respect of the use of prosecution history in claim construction.)

S 194 proposes a complete overhaul of prior user rights under s 56, with the existing section to be replaced entirely. The proposed amendment clarifies prior user rights in several respects, extends the right in some ways and limits it in other ways. It appears that the proposed s 56 was modeled on the UK Patents Act 1977 s 64, though with notable differences. The corresponding provision in US law is 35 USC § 273. A thorough review of Canadian law up to 2001 is provided by Gregor Binkley, Prior User Rights and the Canadian Patent Act, (2001) 18 CIPR 207, and James G Fogo, “The Imperfect Monopoly—The Application and Effect of Section 58 of the Patent Act” (1962), 38 CPR 147 provides an earlier review which is also very helpful, particularly in respect of the history of the provisions.

The current s 56 provides that a prior user has the right to use “the specific article, machine, manufacture or composition of matter” that was made or acquired prior to the claim date. The proposed new section 56(1) provides that if a person “committed an act that would otherwise constitute an infringement of the patent,” it is not an infringement of the patent (or CSP) if the person commits “the same act” after the claim date. The right also arises if the prior user made “serious and effective preparations to commit such an act.”

This reference to “an act” has two effects. First, it evidently also encompasses carrying out a process, a point which was not clear under prior law: Binkley 214-15, 220-224. More generally, there is no restriction on the kind of invention which the “act” might infringe, so it is apparently intended that prior user rights may arise in respect of any kind of invention. Second, it transforms the right from a right to use the specific article, to a right to continue to carry out the infringing act. For example, suppose the patented product is a new kind of toy. Under the current Act, if the prior user had made some infringing toys prior to the claim date, it would be entitled to subsequently sell those specific toys, but not toys manufactured after the claim date. Under the proposed provision, it would be entitled to continue to make new toys.

Gillette Defence (Not Really) Applied

TearLab Corp v I-Med Pharma Inc 2018 FC 164 Manson J
            2,494,540/ TearLab System / i-Pen System

It was lucky for TearLab that its application for an interlocutory injunction was denied by Manson J 2016 FC 606 aff'd 2017 FCA 8 (see here), as TearLab has now lost the infringement action, on the basis that the asserted claims were infringed but invalid as being anticipated and obvious. (Had TearLab been granted the interloctory injunction, it would now be liable to I-Med on the undertaking in damages.) The result turned on a self-created squeeze between infringement and validity:

[183] That result flows directly from the dilemma created by the Plaintiff's urged broad construction for the asserted claims to include both in vivo and ex vivo applications of the invention, which, while I have agreed to that construction, leads to invalidity due to anticipation.

The Gillette defence was invoked, though not really applied. As Manson J explained, the Gillette defence is based on the House of Lords decision in Gillette Safety Razor Co v Anglo-American Trading Co Ltd (1913) 30 RPC 465 (HL), holding that “a defendant may plead that their alleged infringing actions are part of the prior art, and therefore the patent is either invalid for claiming subject matter included in the prior art or, if the patent is valid, the defendant cannot infringe” [179]. The key feature of the Gillette defence is that it obviates the need for the defendant to establish whether the patent is invalid or infringed; once it is established that the defendant’s device was part of the prior art, one or the other must be true, and it is not necessary for the defendant to establish which. Indeed, in principle, it is not even necessary to construe the claims. As Lord Moulton explained at 480-81:

I am, therefore, of opinion that in this case the Defendants' right to succeed can be established without an examination of the terms of the Specification of the Plaintiffs' Letters Patent. I am aware that such a mode of deciding a Patent case is unusual, but from the point of view of the public it is important that this method of viewing their rights should not be overlooked. In practical life it is often the only safeguard to the manufacturer. It is impossible for an ordinary member of the public to keep watch on all the numerous Patents which are taken out and to ascertain the validity and scope of their claims. But he is entitled to feel secure if he knows that that which he is doing differs from that which has been done of old only in non-patentable variations, such as the substitution of mechanical equivalents or changes of material shape or size. The defence that “the alleged infringement was not novel at the date of the plaintiff's Letters Patent” is a good defence in law, and it would sometimes obviate the great length and expense of Patent cases if the defendant could and would put forth his case in this form, and thus spare himself the trouble of demonstrating on which horn of the well-known dilemma the plaintiff had impaled himself, invalidity or non-infringement.

In TearLab, while the Gillette defence was raised, it was not really applied, because Manson J did construe the claims, and held them to be infringed but anticipated (and also obvious). The advantages of the Gillette defence are likely more theoretical than practical, as it would be a bold defendant that would put all its eggs in one basket of proving that its product was the same as the prior art, without bothering to construe the claims, or address validity and infringement directly. It is perhaps worth noting that the Gillette defence was not put forward by the defendant in Gillette itself, but rather was developed by Lord Moulton on his own initiative: 477.

Applying for an NOC Does Not Sustain an Action

Purdue Pharma v Collegium Pharmaceutical, Inc 2018 FC 199 O'Reilly J
            Oxycodone

Purdue Pharma began an infringement action against Collegium for infringement of Purdue’s patent related to oxycodone, on the basis that Collegium had taken various steps to obtain marketing approval in Canada in respect of its oxycodone product. The steps alleged included importing the drug for purposes of its NDS, filing an NDS, obtaining approval in the US, publicly expressing its intention to market the drug in Canada, and invoking the NOC Regulations by serving Purdue with an NOA [21].

Reversing Aalto J, O'Reilly J granted Collegium’s motion to strike. He held that these facts did not allege anything going beyond the regulatory use exemption in s 55.2. There is no requirement that the party seeking to strike on that basis tender evidence that the activity in question did not go beyond attempting to meet regulartory requirement: “Rather, a statement of claim will be deficient on its face if it merely alleges activity that falls within the statutory exemption in s 55.2" [20].

Nor will these facts support a quia timet action. The prospect of Collegium launching at risk after succeeding in the NOC proceeding is not in itself sufficient to satisfy the requirement of imminence, given that it might never actually obtain the NOC prior to expiry of the patent [26]-[28].

More generally, this decision implies that allegations which amount to saying that the defendant was attempting to meet Canadian regulatory requirements will not survive a motion to strike [28]. Note that Purdue has also instituted NOC proceedings, which are not affected [1]. Of course, this will all be moot with the new NOC Regs.

Draft Trans-Pacific Partnership Treaty – Regulatory Review Exception

The leaked TPP treaty contemplates a regulatory review exception in Art E.13, but there is some controversy over the scope. There is considerable debate over the precise wording (which makes the annotated draft difficult to read), but main debate is over whether the exception should be confined to getting marketing authorization for pharmaceuticals, or whether it should extend to any kind of  regulatory approval. The US proposal is representative of the more restrictive position, and the Canadian proposal is representative of the broader position. I have parsed out their positions below, with my underlining to highlight the differences. (No other countries support either of these positions exactly, but several support each of them with some deviations.)

Experimental and Regulatory Use Defence Available on the Facts in Imatinib

Teva Canada Ltd v Novartis AG / imatinib 2013 FC 141, Snider J

            GLEEVEC / imatinib mesylate / 2,093,203

Having successfully established the validity of its patent, Novartis sought a declaration of infringement in respect of certain volumes of bulk imatinib held by Apotex. Snider J held that the defence of regulatory use (s 55.2(1)) or the common law experimental use defence, preserved by s 55.2(6), was applicable in respect of bulk imatinib used for the following purposes [395]:

(a) developing suitable formulations and processes;

(b) obtaining regulatory approval to sell commercial formulations; and

(c) demonstrating that its manufacturing process could be carried out on a commercial scale.

Snider J did not deal with the statutory and common law defences separately, though she ultimately held that both were applicable [401.1]. This result is generally consistent with the cases following Micro Chemicals [1972] SCR 506, especially Merck / lisinopril, 2006 FCA 323 [109-13]. The facts on this case are very similar to those in Merck / lovastatin 2010 FC 1265, [631-32] in which Snider J also held the regulatory or common law defence to be available. As Snider J stated in Servier v Apotex / perindopril, “Of critical importance, in my view, none of the raw material or the actual formulations that were made in the course of that development process were ever sold or used for a commercial purpose” 2008 FC 825 [166] aff’d 2009 FCA 222 [24]. 

The importance of lack of commercial use was emphasized by Snider J’s holding with respect to a certain volume of bulk imatinib held in Apotex’s inventory. The defence was not available for this inventory material, so the material was infringing, but Snider J nonetheless declined to exercise her discretion to order delivery up [401.2]. This was primarily because the evidence indicated that this material would not be used for commercial purposes [397], and, importantly, Dr Sherman gave an formal undertaking, which was incorporated in the judgement, that the inventory bulk imatinib would never be used commercially [399-400].