Purdue Pharma sought to list patent 2,098,738 against TARGIN under the PM(NOC) regulations. TARGIN is a controlled release combination of two active ingredients, oxycodone and naloxone. The ‘738 patent claims a controlled release oxycodone formulation “comprising” oxycodone in a matrix. None of the claims expressly mention naloxone. Does the ‘738 patent encompass TARGIN? If it does not, then a generic version of TARGIN cannot infringe, and the ‘738 patent is not eligible for listing. That question, which turns on the meaning of “comprising”, is the subject of this post. Whether the ‘738 patent would be eligible for listing even if the TARGIN did potentially infringe is a separate question which will be discussed in a subsequent post.
Normally “comprising” defines an open-ended list which does not exclude additional unrecited elements or method steps. On this standard interpretation, a formulation of TARGIN which used the claimed controlled released technology would infringe the ‘738 patent. However, Crampton J accepted the submission of the OPML that in this patent, “comprising” was a limiting term, from which is follows that TARGIN would not infringe.
Crampton J’s reasoning was fatally flawed. He expressly held at [FC 43] that the claims construction for the purpose of the listing requirement could be different than for the purpose of infringement and validity. The FCA held that this was wrong: the claims must be construed in the same way for all purposes [17]. Crampton J was of the view that this divergence was necessary to achieve the objective of product specificity which is central to the listing requirement embodied in section 4 [FC 44]. However, because listing eligibility and infringement are independent, product specificity can be achieved by the application of section 4 itself, without requiring distortion of the claims construction.
This is all sound and logical. Unfortunately, the FCA went on to indicate that even when properly construed, “comprising” in this claim did not denote an open-ended list. The Court held, in effect, that the term “comprising” does not have any presumptive general meaning [23]:
“Reliance on other cases (two were cited) where “comprising” was construed to include a particular feature does not transform the result in those cases to a principle of general application. Cases are rarely identical and frequently are not even similar. Each will turn on its facts. The interpretation accorded “comprising” in the two authorities cited by Purdue was based on the particular patents in suit and the evidence in relation to them. They are not helpful in this matter.”
This is extremely disappointing. It is widely recognized that the ability to determine the scope of a claim ex ante is extremely important to a patent system that promotes rather than impedes innovation: see Chapter 2 of the US Federal Trade Commission Report on "The Evolving IP Marketplace.” Legal rules that improve certainty in claim construction are therefore very important. This is not to say that we should adopt a legalistic approach to interpreting claims in the name of certainty. But words do have meaning, which is why, even on the modern approach to statutory construction, or the purposive approach to claim construction, the text is our starting point. It is true that no term is entirely precise, but it would be absurd to say that the meaning of “enantiomer” or “piston” does not have any general meaning. This is particularly so when we are dealing with a term of art, such as “comprising,” which is used by a claims drafter precisely because it has a generally accepted meaning. It is true that a patentee may make its own dictionary, but when a standard term is used, that dictionary must be very clear in order to change that meaning. If the claim says “black,” then the disclosure would have to be extremely explicit for “black” to be interpreted as meaning “white”. Using “comprising” to indicate a closed list is not very far from using “black” to mean “white”.
In his decision Crampton J held at [FC 46-47] that passages from the patent itself support the view that the claim was to a dosage form containing oxycodone as the sole medicinal ingredient. My own view is that these passages do not at all support that conclusion; but even if they do support that conclusion weakly, the passages cited by Crampton J are certainly not sufficient to reverse the meaning of a standard term.
The other main point relied on by both the Crampton J and the FCA is that to interpret “comprising” as open-ended “would recognize a potentially unlimited number of unnamed other medicinal ingredients to be within the scope of that claim” [26] (see also [FC 49]). With respect, there is nothing to this point. Imagine an entirely new and potent anti-cancer drug has been invented. It is uncontroversial that it would be entirely proper for a claim to that drug to encompass the drug in combination with all the other medicinal and non-medicinal ingredients with which it might be combined, from anti-nausea or anti-constipation drugs, to a wide variety of excipients. This is was clearly established in Burton Parsons, [1976] 1 S.C.R. 555 where “hundreds if not thousands,” of different unnamed compounds might be used. Framing a claim of that type is exactly the function of the word “comprising.” While a broad claim may be invalid because for one reason or another, it is well established that “[t]he breadth of a claim is not per se a ground of invalidity” (Mobil Oil Corp. v. Hercules Canada Inc. 57 C.P.R. (3d) 488 (FCTD).
This unfortunate discussion of the meaning of “comprising” was strictly obiter, as the FCA [27] did not find it necessary to come to a firm conclusion on this point in light of its conclusion on the listing eligibility issue. Nonetheless, this singular focus on context over text is a disturbing portent of its approach to claims construction. Derrida has had his day as a philosopher, but his views seem to live on at the FCA.
No comments:
Post a Comment