Tuesday, August 20, 2019

First Mover Exclusivity for First Generic to Serve an NOA Not Relevant to Procedure under PM(NOC) Regulations

Bayer Inc v Teva Canada Limited 2019 FC 1039 Pentney J
            2,547,113 / 2,624,310 / 2,823,159 / rivaroxaban

This motion to add defendants in a trial of common issues under the PM(NOC) Regulations raises the issue of whether the first generic to serve an NOA can get a practical head start in the market over later generics. Under the US Hatch-Waxman patent linkage system, the first applicant to challenge a drug patent is entitled to 180 days of exclusivity against subsequent generic applicants: 21 U.S.C. § 355(j)(5). The theory is that initiating and defending an action will be expensive for the first generic, but all subsequent generics will benefit at no cost or risk to themselves if the patent at issue is declared invalid with in rem effect. The 180 exclusivity is intended to provide an incentive for the first generic to bear the cost of litigation. There is no equivalent statutory exclusivity period in the NOC Regulations, but the first generic to get its NOC will nonetheless get a practical head start. This means that whether a generic gets any period of exclusivity turns on the procedural details under the Regulations.

The background to this motion is that Teva and Apotex served NOAs on Bayer arising from their applications for an NOC for a generic version of rivaroxaban. These NOAs were served within a month of each other, and Bayer responded by launching actions against Teva and Apotex, again within one month of each other [4]. The claims relate to the same drug, the same patents, and same issues of invalidity [1] and in Bayer v Apotex 2019 FC 191 Tabib J ordered a common hearing of the common issues [2]. A few months later, Taro and then Sandoz also served NOAs on Bayer relating to the same drug, and Bayer commenced actions against them as well [4]. The claims in all four actions are “virtually identical” [34].

In this motion, Bayer, Taro and Sandoz sought to have Taro and Sandoz added as defendants in the trial of common issues already set for Teva and Apotex [1]. Teva and Apotex objected, on the basis, inter alia, that Taro and Sandoz were late in serving their NOAs, and Teva and Apotex should not be required to cede the potential commercial advantage they might gain by having their action decided first. “Adding Taro and Sandoz now as defendants to the common hearing would allow them, in effect, to ‘leapfrog’ the expected sequence of events within the 24-month timeline fixed by the Regulations, and thereby re-gain the commercial advantage they have lost” [7].

Pentney J rejected this argument, agreeing with Tabib J’s remarks at [7] in Bayer v Apotex [13]. Tabib J made two main points. First, the Regulations themselves do not provide any exclusivity period for the first generic to serve an NOA. The US model, including its 180-exclusivity period, must have been considered when the new Regulations were drafted, and as Taro argued, “the Court should not create such a guarantee by its procedural rulings” [15]. Further, as a practical matter, refusing to order a common hearing would not guarantee that the first generic to serve would be the first to come to market.

These are both compelling points. The question remains whether the Canadian system provides an adequate incentive to challenge a patent by serving an NOA. The answer to this question is not clear. While the theory behind the US approach is logical, I’m not sure exactly how different the systems are in practice. I understand that a full 180 day exclusivity can be quite elusive in the US, with the exclusivity often shared among multiple generics; and conversely, in this case Taro and Sandoz are not simply free-riding off the efforts of Teva and Apotex, but will have to bear some of the litigation costs as well. No doubt the US 180 day exclusivity does create some real practical differences, but there is no simple theoretical answer to which system is better.

With this argument rejected, Pentney J decided the matter by reference to the usual considerations of balancing judicial economy against potential prejudice to the parties, and he concluded by adding Taro and Sandoz as defendants in the hearing of common issues.

Wednesday, August 14, 2019

Background Information in the Specifications May Be Evidence of the CGK

Corning Cable Systems LLC v. Canada (Attorney General) 2019 FC 1065 McVeigh J
            2,679,996 / 2,754,149 / Optical Splitter Module

In this decision McVeigh J, applying a reasonableness standard, upheld the Commissioner’s decision to refuse the 996 and 149 applications on the ground of obviousness [107]. The decision applied settled law to the facts.

The applications at issue relate to a “Local Convergence Point” which is a box that receives a cable supplied by an internet service provider as an input and splits and distributes that cable to several units in a building [5]. The claimed invention took advantage of recently developed “bend performance optical fibers,” which can be bent to a smaller radius than traditional fibers without incurring a marked loss in signal quality [6]. This permitted design of a smaller box. As the application explained as part of the “Background of the Invention” “Conventional LCPs . . . require large, expensive LCPs that may be difficult to install and/or transport. In addition, conventional LCPs often require skilled technicians to install the LCP and route the associated subscriber cables. Therefore, a need exists for LCPs that are cost-effective, are relatively small in size, and may be installed and maintained by relatively unskilled technicians” [15]. The Commissioner relied on this statement in finding that identifying the need for a smaller box was common general knowledge [47].

Corning argued that adopting information provided in the background of the patent applications as CGK, without some evidence beyond information in the patent applications themselves, was an error of law [51], [53]. While the decision was not entirely explicit on this point, as I understand it, this issue was important because Corning wanted to argue that identifying the problem – the need for a smaller box – was or contributed to the inventive step necessary to support a patent. I must say that arguing that it was inventive to thinking of trying to shrink the size of electronic equipment strikes me as pretty desperate. In any event, McVeigh J held, correctly in my view, that this point had been settled by Newco Tank 2015 FCA 47, which held, on essentially identical facts, that it was “open to the Board to conclude that the skilled person’s CGK was reasonably described by reference to the language presented as background information in the Patent” [62], quoting Newco at [13] (and see my post on Newco here). McVeigh J did note, however, that “This is not to say that it is always appropriate for the Commissioner to find that information presented in the specifications of a patent application is the CGK of a person skilled in the art. While I disagree with Corning that concluding that such information is CGK – without further evidence – necessarily amounts to a reviewable error, there may certainly be cases in which such a finding is unreasonable” [64].

Corning also argued that the Commissioner erred in finding that actually implementing the smaller box did not require inventive ingenuity. McVeigh J affirmed that this finding was reasonable, noting that “it was insufficient for Corning’s counsel to set forth, during the requisition process and at the hearing, a number of alleged experimental difficulties without concrete evidence” [105].

Friday, August 2, 2019

Should a PAE Be Entitled to Elect an Accounting?

T-Rex Property AB v Pattison Outdoor Advertising Limited Partnership 2019 FC 1004 Manson J
            2,252,973 / Digital Information System

In comparison to the US, we have not seen a lot of litigation by patent assertion entities (PAEs) in Canada, but some actions are now starting to work their way through the system. This decision of Manson J upholding a bifurcation order and affirming the scope of a protective order granted by Steele J, while routine in many ways, raises some interesting points as to how the Canadian litigation system will deal with PAEs.

A first point is that Manson J explicitly referred to the plaintiff, T-Rex Property, as a “non-practicing entity” [27] (NPE). It appears that T-Rex is not just an NPE, but more specifically a patent assertion entity (PAE). The distinction is that an NPE is any entity that does not practice the patent, while a PAE is an entity whose only business is enforcing patents. While PAEs are all NPEs, the converse is not true. Small research based firms and universities that develop new technologies and licence it to others for commercialization are NPEs but are generally not considered PAEs.

A second point is that Manson J noted that “the question of whether a non-practicing entity is entitled to elect between damages or profits appears to be a novel legal issue” [27]. This is interesting simply because Manson J explicitly framed the question as turning on the nature of the plaintiff as a PAE. I think this is the first time I have seen it suggested that a PAE should be treated differently in law, at least in Canada. In US practice “district courts appear to have adopted a de facto rule against injunctive relief for PAEs and other patent owners who do not directly compete in a product market against an infringer” (Seaman, “Permanent Injunctions in Patent Litigation After eBay: An Empirical Study,” 101 Iowa L Rev 1949 at 1953); but doctrinally, at least, this is not a distinction based on status, but is rather because it is more difficult for a PAE to prove irreparable harm in the eBay test.

Apart from the framing, the question of whether an NPE, and particularly a PAE, should be allowed to elect an accounting is very interesting. A successful patentee is entitled to damages as of right, but an accounting is discretionary. Patentees are typically permitted to elect an accounting more or less routinely, but it is not unheard of for an election to be refused. The principles governing whether an accounting should be permitted are quite underdeveloped: it is not even clear whether a successful patentee is presumptively entitled to elect an accounting (see here and here), nor is it clear what factors should be taken into account in the exercise of the court’s discretion. The question of whether a PAE, or an NPE, should be entitled to elect an accounting would not merely be a question of applying established principles, but rather one of developing those principles. Without expressing any view on this difficult issue, I would say that this is an issue where the distinction between a PAE and an NPE is potentially important one.

The procedural points at issue in the motion may also be significant in shaping PAE litigation. The defendant Pattison sought an order bifurcating the damages portion of the action, and T-Rex resisted. Bifurcation orders are commonly granted in patent cases to avoid wasting money and court resources on an issue that will be moot if the patentee loses on the merits. No new points were raised by Manson J in affirming Steele J’s decision to grant the motion. The more interesting question is why T-Rex opposed the bifurcation. Different PAEs have different litigation strategies (see this FTC Study), but one strategy that is used by some PAEs, especially when the case is weak on the merits, is to offer to settle for less than the defendant would need to spend to defend the case. In aid of this approach, some PAEs will try to impose substantial discovery costs on the defendant in order to gain settlement leverage; this strategy can work because discovery costs tend to be asymmetrical, especially when the patentee is a PAE, and so does not have information as to how the invention was developed. A PAE which uses this strategy might want to resist bifurcation in order to increase settlement pressure with discovery related to the damages issue. Perhaps that is what is going on here — though that’s pure speculation, as I don’t know anything about T-Rex apart from what I’ve read in this decision.

Costs are also an important aspect of PAE litigation, and I’ve heard it suggested that we have seen less PAE litigation in Canada because we follow the English rule of costs-in-the-cause. When the PAE’s case is weak on the merits and the PAE is aiming to settle on a litigation cost basis, costs on motions are even more important than any ultimate costs order. In this decision, Steele J required T-Rex to pay costs for the motion immediately, despite T-Rex having posted security for costs [41]. Manson J upheld this “quintessentially discretionary” decision [42], though remarking that “I may not myself have awarded costs payable forthwith in these circumstances” [44]. To the extent that T-Rex might be employing a litigation cost settlement strategy, the order to pay costs on the motion forthwith might help tilt the settlement leverage back in Pattison’s favour (or at least less against Pattison). While Manson J made no reference to T-Rex’s status as a PAE in this part of his decision, it strikes me that costs on motions might be an issue where the PAE status of the patentee is relevant, and Steele J’s order might be justified on that basis, even aside from its discretionary nature.

With all that said, I’m not sure if tailoring procedural rules to PAEs is desirable (though to be clear, I’m not sure it’s undesirable either). PAE litigation strategy highlights aspects of the patent system that arise to some degree in all types of patent litigation, so in principle, sound general rules should suffice. But PAE litigation may also highlight areas where the general rules need to be developed, as with the principles applicable to an election of an accounting.