Utility of CIALIS Patent Upheld on the Facts

Eli Lilly Canada Inc v Mylan Pharmaceuticals ULC (NOC) 2015 FC 17 de Montigny J
            2,226,784 – tadalafil– CIALIS

In this NOC proceeding, de Montigny J held that Lilly had established that its ‘784 patent was valid in the face of Mylan’s allegations of lack of utility and obviousness-type double patenting. The utility issue turned on the facts, and Lilly won comprehensively, so that alternative constructions of the promise of the patent were largely rendered moot. Perhaps the point of most general interest is that de Montigny J rejected the proposition that the word "treatment" implies lack of undue side effects. The obviousness-type double patenting issue raised one novel point of law, namely the appropriate date for assessing obviousness in the context of double patenting. Tomorrow’s post will look at that question. Perhaps the most interesting legal question raised by this case, albeit tangentially, turned on the interpretation of s 27(5), which provides that when a claim is framed in terms of alternatives, each alternative is to be considered a separate claim. A third post will discuss that issue.

The claims at issue were to tadalafil and 3-methyl tadalafil, or a pharmaceutical composition comprising those compounds, for the treatment of ED in men, specifically (Claim 18) by oral administration [18]-[25]. The argument relating to utility turned on the promise of the patent. The contentious issues were whether the patent merely promised the treatment of erectile dysfunction (ED) by use of tadalafil, as Lilly argued, or whether it also included efficacy in oral administration [76], and / or lack of toxicity [87], as argued by Mylan. Some interesting points of law were raised tangentially, but were largely rendered moot by de Montigny J’s holdings on the facts.

A first point is that in Plavix FCA 2013 FCA 186 (blog), the FCA stated that “it should not be taken to have assumed that every patent contains an explicit promise” [50]. In this case, contrary to that statement, de Montigny J clearly did assume that there must be a promise: see eg [71], [76]. The possibility that the patent might not have a promise at all was not even mentioned. I can only assume that this was because the point was not argued by Lilly. Consequently, statements by de Montigny J such as “The promise of a patent is fundamental to the utility analysis and must be ascertained at its outset,” which are questionable since Plavix FCA, cannot be accorded significant precedential value. Moreover, de Montigny J ultimately held that the promise was that tadalafil could be used to treat of ED [82], and that this promise was satisfied on the facts [100], so the fact that he did not consider whether there was any promise at all became moot.

One point of more general interest is that Mylan got caught in an obviousness / utility squeeze on the issue of whether the promise was satisfied, because in the context of obviousness, Mylan’s experts stated that it was obvious that tadalafil would work to treat ED, and de Montigny J took this as evidence that a skilled person could have predicted that it would work [100]. Mylan did not win the other side of the argument either:

[149] The standard for determining whether there is a sound prediction is obviously lower than the standard for determining obviousness. Sound prediction calls for a prima facie reasonable inference, whereas obviousness requires that it be evident an alleged invention will work. This is precisely why it is not inconsistent either to determine that the promise of the ‘784 Patent is soundly predicted but that it was not obvious (to say nothing of the fact that sound prediction is to be determined about a year later in the case at bar).

On the issue of whether the promise includes oral administration, de Montigny J emphasized the principle that any promise must be clear and explicit, citing Celecoxib 2014 FCA 250 [66] (though the principle was first emphasized in Plavix FCA), and on examination of the disclosure he concluded that oral administration was no more than a preferred feature [86]. The interesting point here is that Claim 18 claimed oral administration specifically. In Fournier / fenofibrate (NOC) 2012 FC 741, [126] Zinn J (blogged here) held:

that claimed utility - is clearly and unequivocally expressed by the inventor in the claims of the patent, then that expression ought to be viewed as the promise of the patent. Any statement found elsewhere should be presumed to be a mere statement of advantage unless the inventor clearly and unequivocally states that it is part of the promised utility

See similarly 2014 FC 38 (here). Also related is the statement by the FCA in Celecoxib suggesting that promises should (normally?) apply on a claim-by-claim basis ([87] (blogged here):

Not one case cited by Apotex stands for the proposition that a promise, once made and shown not to have been met, must be construed as invalidating the invention as a whole”

While de Montigny J did consider the fact that Claim 18 claimed oral administration as one factor in concluding that the promise of the patent as a whole did not encompass oral administration [85], this was only as a consideration (and the last one at that), and not an application of a general principle such as statement by Zinn J. It is not clear what to make of this, as de Montigny J did not mention Zinn J’s holding at all, so it may simply be that the point was not argued. In any event, this point was not determinative, as he held that a promise that tadalafil could be administered orally would also have been soundly predicted [111].

Mylan argued that the promise requires the absence of undue side effects, on the basis that a consideration of side effects was implied by the claim term “treatment” [87]. This has wide-ranging implications, given the common use of the word "treatment" in pharmaceutical patent claims. de Montigny J rejected Mylan's argument on this point, largely on the basis that a promise must be clear and explicit [88], but he also specifically rejected the proposition that the term “treatment” implies lack of side effects [92], [114], [115].

Finally, in the context of discussing whether a promise relating to oral administration would be satisfied, de Montigny J specifically accepted and applied the doctrine that the factual basis for sound prediction must be disclosed in the specification [106], though he also affirmed that it is enough that the factual basis is mentioned in the specification – the data itself need not be disclosed [95], [97]. While some relevant evidence was excluded on this basis, there was nonetheless sufficient evidence to establish a sound prediction of utility in oral administration [111]. It is worth noting that the critical evidence on this point was published after the priority date of the ‘784 Patent but before its Canadian filing date [108]. Thus the curious but well-established rule that utility is assessed as of the filing date (2006 FCA 64, [30]) even though non-obviousness and anticipation are assessed as of the claim date, was important in establishing a sound prediction of usefulness in oral administration

There was also a preliminary question was as to the admissibility of various documents filed by Lilly in a European opposition proceeding to a patent relating to VIAGRA (sildenafil) [27]. While de Montigny J held the documents were admissible [27], he did not give them much weight and they did not play a significant role in his reasoning [139]-[140].