Focus of the Obviousness Inquiry Should be on the Solution to the Objective Problem

Apotex Inc v Shire LLC 2021 FCA 52 Rennie JA: de Montigny, Gleason JJA affg 2018 FC 637 Fothergill J

            2,527,646 / lisdexamfetamine [LDX] / VYVANSE / NOC

My previous post discussed some puzzles that I see as arising from Vyvanse. In this post I’ll try to address those puzzles. In Bristol-Myers 2017 FCA 76 [65] Pelletier JA remarked that

It may be helpful to keep in mind that the obviousness analysis asks whether the distance between two points in the development of the art can be bridged by the Skilled Person using only the common general knowledge available to such a person. If so, it is obvious. The first of those points is the state of the prior art at the relevant date. References in the jurisprudence to “the inventive concept”, “the solution taught by the patent”, “what is claimed” or simply “the invention” are attempts to define the second point.

While Pelletier JA went on to note that these terms are often treated as being synonymous [66], there is arguably a distinction to be drawn between a focus on the information disclosed in the specification, and the claimed subject-matter. This has given rise to a debate as to whether the proper focus of the obviousness inquiry—the “second point”—is the inventive concept or the claims. The debate has centred on the second and third steps of the Windsurfing / Pozzoli approach to the obviousness determination, endorsed in Sanofi 2008 SCC 61 [67], which seems to distinguish the inventive concept from what is claimed while at the same time treating them as alternative means of defining the second point. In Ciba 2017 FCA 225 Pelletier JA indicated that the claims were the proper focus of the inquiry. In Vyvanse Rennie JA held that it is inventive concept that is the ultimate the end point for the obviousness inquiry [65]. At the same time, Rennie JA indicated that the inventive concept is the same as the “solution taught by the patent,” as that term was used in Bristol-Myers. In this post I will argue that Bristol-Meyers holds the key, though it is somewhat buried. The proper focus of the inquiry is not the inventive concept, nor is it the essential elements of the claim; it is whether the subject-matter defined by the claim is the solution to the objective problem that would have faced the skilled person.

There is certainly a powerful and well-founded intuition that the focus of the inquiry should be on the inventive concept. I’ll start off by making the case in favour of the inventive concept, and then I’ll show where that approach breaks down. My next post will look at where this leaves the Windsurfing test.

A New Framework for Construing the Inventive Concept

Apotex Inc v Shire LLC 2021 FCA 52 Rennie JA: de Montigny, Gleason JJA affg 2018 FC 637 Fothergill J

            2,527,646 / lisdexamfetamine [LDX] / VYVANSE / NOC

In Sanofi 2008 SCC 61 [67], the SCC endorsed the four-step Windsurfing / Pozzoli approach to the obviousness determination. The second step of this inquiry is to “Identify the inventive concept of the claim in question or if that cannot readily be done, construe it.” This initially led to a focus on identifying the inventive concept as a central part of the obviousness analysis. Unfortunately, with no clear test or methodology, identifying the inventive concept has proven to often be difficult, as the FCA noted in Bristol-Myers 2017 FCA 76 (see here). In Ciba 2017 FCA 225 the FCA remarked that the search for the inventive concept can be a “distraction” [76] that has resulted in “confusion” and “uncertainty” that can be reduced “by simply avoiding the inventive concept altogether and pursuing the alternate course of construing the claim” [77]: see here. Consequently, the recent decisions of the FCA “have downplayed the importance of the ‘inventive concept’ as an analytical tool in the context of an obviousness analysis,” and have instead focused the analysis “on the claims themselves”: TearLab 2019 FCA 179 [78]. In Vyvanse the pendulum has swung back towards a renewed emphasis on the inventive concept, with Rennie JA stating that “The inventive concept, properly construed and applied, remains the end point for the obviousness inquiry” [65].

It is not very satisfactory to imagine the jurisprudence of the Court swinging back and forth with the needs of the moment. Bristol-Myers, Ciba, and Vyvanse make sense in their own terms; the more difficult question is how to reconcile them. In this post I will focus on Vyvanse, which is straightforward in some respects but puzzling in others. In the next post I will argue that an approach to obviousness that focuses on the solution to the objective problem faced by the skilled person is the only way to reconcile Bristol-Myers, Ciba and Vyvanse. In a third post, I will argue that the Windsurfing approach is useless or worse and should be abandoned.

Leave Granted in Nova v Dow

Nova Chemicals Corporation v Dow Chemicals Company 2020 FCA 141 Stratas JA: Near JA / Woods JJA dissenting, affg 2017 FC 350, 2017 FC 637 Fothergill J

            2,160,705 / film-grade polymers / ELITE, SURPASS

The SCC has granted leave to appeal in Nova v Dow. I’ve written several posts on the FCA decision:

Nova v Dow: A Radical Departure from Established Law

Constructing the “But For” World Is Not a Purely Subjective Inquiry

Non-Infringing Baseline as an Alternative to “But For” Causation

Deduction of Fixed Costs in Nova v Dow

Fixed Costs and Sunk Costs

Hat tip to Alan Macek’s IPPractice daily email service.

Empirical Evidence on the Effect of Fee-Shifting

Today I’m pleased to present a guest post:

By Christian Helmers, Yassine Lefouili, Brian J. Love, Luke McDonagh.

As previously covered on this blog (here, here, and here, for example), changes are afoot for Canada’s costs shifting regime. In 2016, the Federal Rules Committee determined that costs awarded pursuant to Tariff B of the Federal Courts Rules were inadequate, and the Rules Committee is actively working on a set of proposed amendments for publication and approval: see Allergan v. Sandoz, 2021 FC 186, [28].

In a new paper just published in the American Law & Economics Review, we present a theoretical model and report empirical findings that may help inform the Committee’s work. Our paper examines the effect of a 2010-2013 court reform process limiting the size of costs awards in cases brought in the UK’s Intellectual Property Enterprise Court (IPEC, formerly known as the Patents County Court), a venue that specializes in hearing small-to-mid-size IP disputes.

Importantly for our analysis, these reforms did not extend to the High Court of England and Wales, the UK’s separate venue for relatively large IP cases. Pre-reform, costs were awarded in both jurisdictions under a common set of principles that allowed victorious litigants to request recovery of their actual costs and recoup on average about half to two-thirds of that amount. Post-reform, while costs awards in the High Court carried on as usual, costs awards in IPEC cases were capped at a maximum of £50,000, restricting the extent to which costs are shifted from losing to winning parties.

To study the effect of this costs cap, we hand-collected data from physical case files for all available IP suits that were filed in either court from 2007 to 2013. We use this data to make two comparisons. First, we compare cases filed in the IPEC to cases filed in the High Court, in effect using High Court cases as a control group to isolate the reform’s effect on IPEC cases. Second, because a lump sum costs cap is likely to have a greater effect on (relatively high cost) patent cases, we additionally compare patent and non-patent IP cases litigated in the IPEC.

Our empirical results suggest that the reform had a sizable impact. First, the rate of IP litigation increased once the costs cap came into effect. We find that patent case filings increased in both comparisons (IPEC to High Court, and patent IPEC to non-patent IPEC) and for both small and large plaintiffs. In addition, with the costs cap in place, we see evidence that cases settled more often and that plaintiffs both won less often and spent less on litigation. Specifically, we find that, post-reform, small IP enforcers won less often in cases decided by the IPEC, that IPEC cases settled at a higher rate, and finally that large plaintiffs retained smaller (and thus presumably less costly) law firms when litigating in the IPEC.

What do these findings portend for a potential increase in average costs awards in Canada? While we study a transition from higher to lower average costs awards, the Canadian Federal Court is in the process of doing the opposite. Accordingly, it seems reasonable to hypothesize that the Federal Court might experience effects opposite to those of a costs cap: i.e., a reduction in the rate of IP litigation, along with a higher plaintiff win rate, lower settlement rate, and increase in litigation expenditures among cases that are filed.

Whether or not these effects (individually or collectively) are normatively desirable is another question, and one with no easy answer. While the UK reform that we study has generally been regarded as a success, costs shifting rules have a notoriously complex impact on litigants’ incentives. In the paper, we ultimately take no normative position on the desirability of a costs cap, and so here too we leave it for the Rules Committee’s consideration whether the potential effects identified above align with the broader goals motivating its work.

Thirty Day Listing Period under NOC Regs Is Not Discretionary

Merck Canada Inc v Canada (Health) 2021 FC 345 Fothergill J

             2,830,806 / pembrolizumab / KEYTRUDA

Subsection 4(6) of the PM(NOC) Regulations provides that in order to be listed on the Patent Register, the patent list must be submitted within 30 days after issuance. The 806 patent was issued on May 12, 2020 [7]. It would otherwise have been eligible for listing against KEYTRUDA, but due to an apparent oversight, the 806 patent was not submitted until after the close of business on June 12, 2020, and was consequently deemed to have been filed on the next business day, Monday, June 15, 2020 [8], which was just outside 30 days. (The 30^th day after issuance was the June 11, if I’m counting correctly.) Tthe Minister consequently refused to list the 806 patent on the Register on the basis that the list had not been submitted within 30 days of the issuance.

Merck appealed on the basis that the Minister’s determination that the Covid-19 Time Limits Act did not have the effect of suspending the 30 day time period specified in s 4(6) was unreasonable, and that the Minister’s determination that she had no discretion to extend the 30 day time period was also unreasonable [14]. Fothergill J found that the Minister’s determination on both these points was reasonable. I won’t go into the details except to say that, for what it’s worth, Fothergill J’s reasoning strikes me as sound.

Open Question on Applicability of Discoverability Rule to Patent Act Limitations Period

Google Canada Corporation v Paid Search Engine Tools, LLC 2021 FCA 63 de Montigny JA: Webb, Rennie JJA affg 2020 FC 992 McVeigh J

            2,415,167

This is an appeal of a order dismissing Google’s motion for a determination of a question of law pursuant to Rule 220(1)(a). The underlying action was brought by Paid Search against Google alleging infringement of its 167 patent. The patent was published in 2002 and granted in 2017, so a considerable portion of Paid Search’s claim would be for “reasonable compensation” for pre-issuance infringement pursuant to s 55(2). The question proposed by Google was whether the six-year limitation period in s 55.01 applies “such that in the present action no remedy, including no reasonable compensation, may be awarded for any act of infringement committed prior to January 12, 2012?” [5]. At first instance McVeigh J dismissed the motion. The FCA has now affirmed the result, but with a very important difference in the reasoning. In particular, as discussed here, McVeigh J had held that the discoverability rule does apply to the s 55.01 limitation period. de Montigny JA’s analysis did not require him to address this question at all, and consequently [12]:

I wish to express no view as to the substance of the legal question raised by the appellants on their motion, and these reasons should not be taken as an endorsement of the views expressed by the Motion Judge as to the application of the discoverability principle in the context of section 55.01 of the Patent Act.

Whether the discoverability rule applies to the Patent Act limitation period therefore remains an open question.

Rule 220(1)(a) provides that “A party may bring a motion before trial to request that the Court determine a question of law that may be relevant to an action.” This involves a two-step test: “the Court must first determine whether it is appropriate, in the specific circumstances of the case, for the proposed question to be addressed in a preliminary motion before trial. It is only when the Court answers that question affirmatively that it proceeds to the second stage to determine the legal question as set out and approved by the Court at the first stage” [6]. The first stage—whether court should address the question—itself involves three requirements: (1) that there is no dispute as to any material fact; (2) what is to be determined is a pure question of law; and (3) its determination will “eliminate the necessity of a trial or, at least, shorten or expedite the trial” [7].

In her decision at first instance, McVeigh J erred by conflating the two questions. She held that the discoverability rule applies to the limitation period in s 55.01, and consequently, the question at issue was not a pure question of law, and for that reason she dismissed the motion [11].

The FCA held that the proposed question was indeed a pure question of law that is not predicated on the assessment of facts [16]. But that only addresses parts (1) and (2) of the first stage. At the step (3), de Montigny JA held that Google had not established that there would be any substantial saving of time and expense at trial, partly because Google had not adduced sufficient evidence in support, partly because the trial preparations were already well underway, and partly because quantification of the remedy would still be required for the time within the limitation period, even on Google’s view of the effect of s 55.01 [21]–[26]. Thus, the FCA affirmed, but on the basis that the question was not suitable for determination on a Rule 220(1)(a) motion; consequently, there was no need to address the question of law itself.

One other point is worth noting. In Jay-Lor 2007 FC 358 [122] Snider J rejected the view that “reasonable compensation” under s 55(2) means the same as s 55(1) damages, and, instead equated it to a “reasonable royalty,” in the absence of any other alternatives presented to her. In Dow v Nova 2017 FC 350 Fothergill J held, in effect, that in some circumstances lost profits could be awarded under s 55(2): see here. In this decision, de Montigny JA agreed that when reasonable compensation is claimed under s 55(2), the remedy “is often” in terms of reasonable royalty” [23]. I take this careful language as signalling that the FCA considers the question of whether reasonable compensation is the same as a reasonable royalty to remain an open question.

Routine Correction of Inventorship

CAE Inc v Canada (Commissioner of Patents) 2021 FC 307 Manson J

            3,000,463

This was a routine correction of inventorship under s 52. Two of the four inventors listed on the priority US application had inadvertently been left off the Canadian application [3]–[5]. The application was supported by affidavits and it was established that the mistake arose through inadvertence. No interesting issues were raised and the application for correction was granted [5].

Claim Construction and Inducement Potpourri

Guest Tek Interactive Entertainment Ltd v Nomadix Inc 2021 FC 276 McHaffie J

2,600,760 / 2,750,345

In Guest Tek v Nomadix McHaffie J held Guest Tek’s 760 and 345 patents to be valid but not infringed [284], [437], in a very careful—and long–decision that turned almost entirely on claim construction and the facts. One thing that jumped out is a stylistic innovation, namely the use of italics to indicate claim elements: for example “In my view, the POSITA would understand the notion of user load to be a function of the number of current users and/or their demand” [340]. I love it. For whatever reason, the use of italics rather than more traditional quotation makes the sentences flow more naturally while still retaining all the same information. It also means that quotation marks are reserved for quotations from the evidence, which eliminates ambiguity when a paragraph quotes the experts about the meaning of specific claim terms.

Turning to the substance, Guest Tek and Nomadix each provide network gateway equipment and supporting software to hotels, which allows multiple hotel guests to access the internet through the gateway provided by the hotel [14]. The patented technology relates to network security and bandwidth management. I must say that after reading pharma cases for years, I found the technology at issue in this case a bit difficult to follow. McHaffie J’s decision generally simply applied established law, apart from the points discussed in Friday’s post, but a few points of some interest were raised in respect of claim construction and inducing infringement. The validity attacks on both patents do not appear to have been pursued very vigorously. It seems that Nomadix’s stronger arguments were addressed to the scenario where Guest Tek had prevailed on claim construction and infringement.

Claim Construction

Friday’s post discussed McHaffie J’s conclusion that recourse to the disclosure is always permissible in claim construction, not just when the claim term is ambiguous when read in isolation. However, this holding did not ultimately affect the analysis: there were two cases where reference to the disclosure was necessary to resolve an ambiguity, and reference to the disclosure otherwise simply confirmed the construction arrived at from the claims alone [48].

McHaffie J’s analysis otherwise applied established principles to the facts, but there are a few points of passing interest.

Claim element redundancy

McHaffie J noted that the principle against redundancy is typically applied as between claims (ie, to avoid claim redundancy) “but in my view it has some value in assessing redundancy of elements within a claim” [150]. I don’t think I’ve seen that point made before; it makes sense to me. On the facts, he nonetheless ultimately preferred a reasoning that resulted in claim element redundancy [150]; this is consistent with the general point that the principle against redundancy is an aid to construction and not a strict rule of law.

“First” and “second” distinguish rather than sequence

Nomadix had argued that the reference to “first packets” and “second packets” were used to indicate the sequence of in which the packets were sent. McHaffie J rejected this, holding that the cases “confirm that the terms ‘first’ and ‘second’ are used in the practice of patent drafting as a means of distinguishing elements rather than necessarily sequencing them” [100]. This is not very surprising (and of course there may be patents where the terms are used to indicate sequence), but McHaffie J’s discussion is a good citation on this point if one is needed.

Constructions making infringement impossible are to be avoided

McHaffie J held that a claim should not normally be construed so that no real-world product could possibly infringe [133], [135].

Balance between text and purpose

McHaffie J noted that claim construction requires “a balance between two evocative terms well known in patent law: the Court should construe a patent with ‘judicial anxiety’ to support a useful invention, but if the inventor has created troublesome limitations in the claims, this is a ‘self-inflicted wound’ the Court will not cure” [50]. The decision provided an nice illustration of this balance; at one point McHaffie J adopted a purposive construction, looking to a function of the device to clarify the meaning of an ambiguous term [188], but at the same time, when another element was clear on the text, he hewed to it, even though that might have resulted in a self-inflicted wound [166]– [167].

Inducement

Guest Tek argued that Nomdix induced infringement of the 760 by hotels in Canada, and did not plead direct infringement [202]. The established law of inducement has a three-prong test Corlac 2011 FCA 228 [162], quoted at [56]:

First, the act of infringement must have been completed by the direct infringer. Second, the completion of the acts of infringement must be influenced by the acts of the alleged inducer to the point that, without the influence, direct infringement would not take place. Third, the influence must knowingly be exercised by the inducer, that is, the inducer knows that this influence will result in the completion of the act of infringement.

Ultimately, the inducement argument failed on the first step, as Guest Tek did not establish direct infringement by any Canadian hotel using a Nomadix gateway [255]. Guest Tek argued that direct infringement occurred when a Nomadix gateway is connected to a wireless access point in a network operated by a hotel, but Guest Tek did not call any witnesses from Canadian hotels, or from Nomadix resellers in Canada, to speak to network configuration in Canadian hotels. Instead, Guest Tek relied primarily on the functionality of the Nomadix gateway itself [204]. Ultimately, as I understand it, Guest Tek at most established that it was possible to configure a Nomadix device so that it would infringe, but there was no evidence that it had actually been configured to infringe in any hotel in Canada: see eg [227], [229]. (There are however, a couple of paragraphs that might suggest that the Nomadix gateway could not be configured to infringe: see eg [232], [234]. I’m sure the point would be clear if I understood the technology a bit better.) In any event, whether or not it could be configured to infringe, non-infringing configurations are certainly possible and apparently quite normal. This conclusion followed in part from claim construction; presumably Guest Tek would have had a stronger argument on direct infringement based on functionality alone if it had prevailed on some of the key claim construction issues. (Though I have to admit I didn’t go back to reverse engineer the arguments to be sure.)

Test for causation at the second branch

McHaffie J accepted that the second branch requires a “but for” test for inducement [58]. While that is certainly the standard articulation of the test, in a draft article, “Is 'But For' Causation Necessary to Establish Inducement?” available on SSRN, I have argued that a material contribution test for causation should be used at the second stage, though “but for” causation is still required to establish damages. Given McHaffie J’s holding on direct infringement, this point did not arise. Moreover, McHaffie J held that Guest Tek “has not established that Nomadix influenced Canadian hotels to configure their wireless access nodes in this way or any particular way” [245], [247], [248]. Consequently, inducement would not be established even on the material contribution test that I proposed in my article; that test comes into play, for example, where the defendant had clearly instructed the direct infringer to use the product so as to infringe, but it is not clear whether the direct infringer paid any attention to those instructions. That is, my argument is that inducement sufficient to support injunctive relief should be established if the indirect party sold a product along with instructions to infringe, and the direct infringer did in fact infringe, whether or not the direct infringer read the instructions. The effect would be that an indirect party could sell a product that could be used to infringe, but would not be allowed to provide instructions to infringe. Damages would still require “but for” causation, not as a matter of the test for inducement, but as a matter of the test for damages.

Contributory infringement

McHaffie J was also kind enough to refer to my article Contributory Infringement in Canadian Law (2020) 35 CIPR 10 (draft version here). In that article, I noted that Canadian courts have consistently expressly said that there is no liability for contributory infringement—see eg MacLennan v Produits Gilbert 2008 FCA 35 [33], but I argued that nonetheless, the reasoning and results in almost all the leading cases support a rule that supply of a product especially adapted to infringe, and with no substantial non-infringing use, constitutes indirect infringement, even in the absence of active inducement. (The exception is Slater Steel (1968), 55 CPR 61 (Ex Ct), which I argued was wrongly decided). McHaffie J remarked while my article was interesting, he agreed with Nomadix “that the FCA has confirmed ‘contributory infringement’ is not recognized as a cause of action in Canadian law,” citing Nycomed 2012 FCA 195 [3]. Now that’s certainly true in terms of express statements of the law; my argument is that it would be open to the FCA to recognize contributory infringement in light of the reasoning and results in the leading cases; but given the express statements of the law, such as that in Nycomed, it would probably be up to the FCA to recognize the point.

In any event, the issue didn’t arise in this case. The issue of contributory infringement arises when the product sold is not in itself patented, but is especially adapted to infringe; the question is whether indirect infringement can be established by mere sale of a product that “cannot be used for any other purpose,” as the FCA put it in MacLennan, even in the absence of a separate act of inducement. As noted above, in this case the inducement allegation failed because Guest Tek did not even establish direct infringement: “As the evidence does not establish Nomadix influenced Canadian hotels with respect to these issues, it equally does not establish any direct infringement would not have occurred but for that influence” [250]. The Nomadix gateway was clearly not a product especially adapted to infringe, or that could not be used for any other purpose; non-infringing configurations were possible and normal. Therefore the issue of contributory infringement does not arise, and it would have made no difference to the outcome or the reasoning if McHaffie J had accepted my argument.