Apotex Inc v Alcon Canada Inc 2016 FC 720 Tabib J
NOC proceedings and infringement actions give rise to independent litigation over infringement and validity of the same patents, with potentially conflicting monetary remedies. This decision by Tabib J on a motion for bifurcation of a s 8 damages action deals with some of the complex procedural issues that arise in attempting to resolve this conflict. While appropriate procedural decisions may help reduce duplicative litigation and avoid inconsistent results, it is hard to avoid the conclusion that the remedial overlap is another illustration that the PM(NOC) Regulations’ well-intentioned attempt to create a simple and streamlined patent linkage system has not been a success.
The patent linkage systems operating in the US and Canada are functionally equivalent to giving an automatic interlocutory injunction to a pharmaceutical patentee threatened by generic entry. In a jurisdiction in which interlocutory injunctions are relatively easy to get, linkage is hardly necessary. So, as I understand it, in the UK a pharmaceutical patentee will normally be able to get an interlocutory injunction against a generic which seeks to launch at risk, and consequently the generic will normally bring a declaratory action to “clear the way.” The effect is that the generic cannot launch before it has been determined that the patents reading on the product are invalid or not infringed. In jurisdictions such as Canada and the US, where interlocutory injunctions are more difficult to obtain, the generic might be able to launch even though its product infringed a valid patent. The patent linkage system addresses this by effectively giving the patentee an automatic interlocutory injunction. In the US, the application for marketing authorization is a deemed infringement, giving rise to a statutory stay which is roughly equivalent to an interlocutory injunction. However, there is no separate hearing to assess the merits of this “injunction,” as there would be in an application for an interlocutory injunction as such. In the Canadian patent linkage system, embodied in the PM(NOC) Regulations, on applying for marketing authorization for a generic drug, the generic is subject to a statutory stay – the automatic interlocutory injunction – while validity and infringement are assessed in an NOC proceeding. The NOC proceeding is intended to be summary in nature, like the hearing in which a true interlocutory injunction application would be decided. And if the patentee is unsuccessful in the proceeding, under s 8 it must compensate the generic which has been wrongly kept out of the market during the period of the stay, just as a party who is granted an interlocutory injunction must normally give an undertaking in damages to compensate the defendant which has been prevented from exercising its legitimate rights. And just as with a traditional interlocutory injunction application, a subsequent infringement action can proceed to a potentially different result on the basis of a more complete record. The Canadian system is therefore intended to statutorily mimic a true interlocutory injunction application more closely than the US linkage system.
However, one feature distinguishes the Canadian linkage system from both a true interlocutory injunction application and the US-style linkage system, namely that the NOC proceeding and any subsequent infringement action are independent proceedings. This means that inconsistent results are possible, in principle and in practice. In the both the US and the UK, a patentee may be liable for foregone profits incurred by a successful generic which was kept off the market by the stay / interlocutory injunction, but in both systems the generic is entitled to compensation only if it prevails in the ultimate infringement action. Thus there can never be a conflict in which the generic is entitled to compensation for having been kept off the market by a stay based on a patent which was valid and infringed. In contrast, in Canada the generic may win the NOC proceeding, and so be entitled to compensation under s 8, and yet lose the infringement action, so that it would be entitled to 8 compensation under s 8 of the NOC Regulations for having been kept out of a market which, under the Patent Act, it had no right to enter.
This potential for conflict has begun to be addressed substantively. As Tabib J states “[t]he law as it stands is to the effect that hypothetical infringement is not a complete defence to a section 8 claim, but that it is a significant factor to be considered in assessing compensation, and that it can indeed reduce damages to zero” (citing 2012 FC 620, blogged here; and see also 2012 FC 559 aff’d 2013 FCA 77 blogged here and here), though as she notes the law “may continue to evolve” .
This decision by Tabib J on Apotex’s motion for bifurcation wrestles with the problem of how to deal with this issue procedurally. Alcon held two patents related to travoprost, the ‘287 and ‘370 patents. Alcon lost both NOC proceedings, and, after Apotex launched, Alcon commenced an infringement action based on the ‘370 patent only. At the same time, Apotex brought an action for s 8 damages, which Alcon sought to defend by alleging infringement of both the ‘370 patent and the ‘287 patent, as well as the ‘172 patent, which was not listed on the Patent Register and so was not at issue in the NOC proceedings , . Apotex now seeks to bifurcate the s 8 action to have validity and infringement of the ‘287 and ‘172 patents determined separately. The parties agreed that validity of and infringement regardng the ‘370 patent were already effectively bifurcated as they will be determined in the infringement action and those findings would be binding in the s 8 action . In some ways the bifurcation decision was quite easy. On the facts, Tabib J found that the issue proposed to be bifurcated were quite distinct, so bifurcation would not lead to wasteful duplication . On the other hand, if bifurcation were refused, duplicative litigation of quantification between the infringement and liability phases might result . Copnsequently, Tabib J granted the motion to bifurcate.
Two interesting points were raised. First, Alcon was apparently concerned that Apotex’s motion for bifurcation was a prelude to a move to consolidate the bifurcated infringement issues relating to the ‘287 and ‘172 patents with the infringement trial of the ‘370 patent, thus delaying the trial and Alcon’s hoped for injunction .While Tabib J gave some credence to this concern , in the end she considered that in this case bifurcation does not entail joinder, and so Alcon would not be disadvantaged in this manner . The concern nonetheless illustrates the complex procedural considerations that arise because of the separate nature of these proceedings.
Secondly and more fundamentally, Tabib J suggested that if Alcon’s related infringement action was not already bifurcated, the best way forward might have been to consolidate both actions, without bifurcation . This would solve the problem of the separate actions in the most straightforward way possible. However, as this case illustrates, consolidation is not always appropriate. And while consolidation might in some cases be possible, given the vagaries of timing of two independent actions, it is too much hope that this would be routine. It seems, therefore, that parallel s 8 and infringement actions will be coordinated only by ad hoc motions such as this one, with varying degrees of success in avoiding wasteful duplication. The procedural complexity created by our patent linkage system seems destined to endure so long as the system remains in its current form.