Wednesday, August 24, 2011

Promise of the Patent Determined Without Reference to the Description

Pfizer Canada Inc. v. Apotex Inc. / latanoprost (NOC) 2011 FCA 236 Trudel JA; Sharlow, Stratas JJA rev’g 2010 FC 447 Heneghan J 

There is not much new law in the FCA’s latest latanoprost decision, but it is significant nonetheless for an extremely aggressive application of the false promise doctrine; the FCA construed the promise of the patent without any reference to the description, and, on the basis of this construction, reversed the Applications Judge and held the claims at issue invalid.

The key claims in the patent at issue, 1,339,132, were a compound claim for latanoprost, a prostaglandin derivative, and a claim for the use of latanoprost in the treatment of glaucoma or ocular hypertension. Prostaglandins had previously been of interest in treating glaucoma, but most (or all) of the prior art prostaglandins were of limited practical use for that purpose because of side effects, primarily ocular irritation.

A crucial issue on appeal was the construction of the promise of the patent. According to standard Canadian law, the utility of the invention is measured by the promise of the patent. This means that even if the invention is sufficiently useful to warrant a patent, the patent will nonetheless be invalid for lack of utility if it promises more than it delivers. At the first instance, Apotex’s argument, which was rejected by Heneghan J, was that the patent promised that side effects would be eliminated or substantially reduced in chronic treatment of glaucoma [FC 69, 192]. The issue was significant because the inventors had only conducted “single dose” studies of eye irritation caused by latanoprost. If the patent promised reduced irritation when given as a single dose, then, as Heneghan J held, demonstrated utility was established. This was not disputed on appeal. Thus it is clear that the invention had that “scintilla” of utility that is accepted as being adequate to satisfy the utility requirement if no higher degree of utility is promised. But if, as Apotex argued, the patent promised that irritation would be reduced even with chronic use, utility depended on sound prediction. The Court of Appeal agreed with Apotex as to the promise of the patent, and in the second issue on appeal, went on to hold that the factual basis for a sound prediction was not established by the single dose studies, and consequently the claims in question were invalid.

So, validity turned on the construction of the promise of the patent, and specifically whether the patent promised that irritation would be minimal even with chronic use. The FCA reviewed the question for correctness as a matter of law [17], and having determined that Heneghan J erred in her interpretation of the promise, came to its own conclusion based on the record, without remitting the matter to Heneghan J [20].

In its analysis the FCA pointed out that the evidence was clear that at the time of filing glaucoma would have been known by the POSITA to be a chronic condition that required chronic treatment [24], and stated that it was “an error to construe the promise of the ‘132 patent without consideration as to the nature of the disease it purports to treat effectively” [27]. Even accepting this, it is not the end of the matter; consideration of the nature of the disease is important only because the POSITA must read the patent with that in mind. The promise of the patent, like the claims, must be interpreted purposively, with attention to text, context and purpose. Certainly, the nature of the disease is part of the context, but that context cannot be divorced from the text. Remarkably, the FCA did not make any reference whatsoever to the patent description in construing the promise of the patent. The entire point was disposed of in 5 paragraphs, [24]-[28], all of which were devoted to establishing that the POSITA would have known that glaucoma was a chronic condition.

If one does read the specification, including the description, the matter is not so clear. Four of the five claims at issue were use claims for a composition for “treatment” of glaucoma [8]. Since glaucoma is chronic, it might be said that “treatment” of glaucoma must be effective against it as a chronic condition. On the other hand, it might be argued that short term relief is a treatment of sorts. In any event, the remaining claim 19 is a compound claim per se, so this argument from the claim language is inapplicable. Turning to the description, I suggest the FCA was off the mark in saying that it is an error to construe the promise of the patent “without consideration as to the nature of the disease it purports to treat effectively.” This formulation begs the question by supposing that the patent purports to treat the disease effectively. Even ignoring that final phrase, the nature of the disease is certainly relevant, but it is only part of the context, as any particular drug may be aimed at any of a number of aspects of the disease. It is more precise, and avoids question-begging, to say that the promise of the patent should be construed in light of the problem that the patent purports to solve. The disclosure of the ‘132 patent expressly describes the problem to which it has provided a solution at 3-4, in terms such as the “irritation experienced when [prior art compounds] are applied.” This implies that the patent promises that irritation will be reduced by latanoprost in the same circumstances in which it was caused by the prior art compounds. If the prior art compounds caused irritation only after chronic use, then this would strongly imply that the patent promised reduced irritation with long term use. But if the prior art compounds caused immediate irritation, then the patent might only promise reduced irritation in the immediate reaction. I don’t know whether the prior art compounds caused immediate irritation, or only long term irritation, but I would argue that this is a central question on the construction of the patent’s promise to reduce irritation. Presumably a POSITA would know, and if the case had been remitted to the trial judge the point could have been settled.

Another point in the disclosure that suggests that the patent only promises immediate relief is in the discussion of the studies (at 18 - 22). As the FCA pointed out, all of these studies were single dose studies. Yet the patent speaks of these studies as showing “absence of any signs of discomfort,” and “complete absence of hyperemia” (18) and “complete loss of ocular irritation” (19), and so on. In this context it is perfectly clear that the absence of irritation refers to absence of short term irritation, because the studies were all single dose. It is a normal presumption that the author of a document uses the same term in the same sense throughout. This suggests that the when the patent at page 4 promised to a solution to the problem of irritation, it meant immediate irritation.

My point here is not simply to argue that the FCA’s interpretation of the promise of this particular patent was wrong. It is possible that the Court was of the view that the reference in the claims to “treatment” was determinative (though that would leave a question regarding claim 19). Even apart from that, I would want to know more about the nature of the problem in the prior art before coming to a conclusion. I have gone through these details show that the FCA established its view of the promise of the patent without making any reference to what is said in the description, and without taking into account what I have argued is important factual context information. It is for these reasons that I regard this as a very aggressive use of the false promise doctrine. Recall that at the Court of Appeal level it was undisputed that the use of latanaprost for the treatment of glaucoma was a patentable invention; the problem was not that Pfizer did not have an invention worthy of a patent, but that in the view of the FCA, it had promised more than it delivered. The Supreme Court has affirmed many times that a patent should be approached "with a judicial anxiety to support a really useful invention": see e.g Consolboard [1981] 1 SCR 504 at 521; Wellcome / AZT 2002 SCC 77 at [92]. I suggest that invalidating an otherwise valid patent for promising more than it delivers, without making any reference to the text of the patent itself, is not consistent with this exhortation. I have argued elsewhere that the false promise doctrine should be abandoned entirely; but if Canadian law is to retain the false promise doctrine to strike down an otherwise valid patent, it should be with more attention to the patent itself than the FCA displayed in this decision.

I do not know how fully arguments related to text of the disclosure were argued on appeal, but in the absence of an outright concession by the patentee that the text was irrelevant, I cannot see how a proper construction of the promise of the patent can be carried out without reference to the patent itself. It is possible in theory that the FCA did consider the description, or that it viewed the word “treatment” as determinative, and simply did not mention its reasoning in this respect. If so, the FCA has failed to give adequate reasons. It has often discussed the duty to give reasons in the administrative law context (eg 2010 FCA 158 at [16]), and all of the justifications for this duty apply equally to a judicial decision. In particular, if patents are to be struck down on the basis of the construction of the promise of the patent, transparency requires that patent applicants be informed of the reasons so they can draft the disclosure in a way that preserves the validity of the claims to a patentable invention. Whether the FCA failed to consider the disclosure in construing the promise of the patent, or simply failed to describe its reasoning, the decision is not at all satisfactory.

My thanks to Alan Macek and his very useful site for bringing this case to my attention before it was released on the FCA website.

No comments:

Post a Comment