2,379,329 / TRIFEXIS
In Gilead / COMPLERA 2012 FCA 254 the FCA apparently held that a perfect literal match between the claims of a patent and the compounds set out in an NOC for a product is required before the patent can be listed on against that product on the Patent Register. In Lilly / TRIFEXISFC Bédard J, following Gilead, refused to allow the 329 patent to be listed against TRIFEXIS. The FCA has now reinterpreted Gilead and consequently set aside Bédard J’s decision. While this FCA decision evidently abolishes the perfect match requirement, it is not entirely clear what the new test is.
The facts in Lilly / TRIFEXIS are fairly simple. TRIFEXIS is authorized as an oral dosage form of a drug that contains two active medicinal ingredients: spinosad and milbemycin oxime. The 329 patent claims an “oral formulation” comprising spinosad [9]. “Oral formulation” was defined in the disclosure to include spinosad in combination with other compounds including “milbemycins” [9]. Consequently Bédard J construed the relevant claims to be directed "not only to a formulation including spinosad as the only active ingredient, but also to formulations that includes other active ingredients such as, but not restricted to, milbemycin oxime" [FC 69]. Thus she in effect held that a generic version of TRIFEXIS would necessarily infringe the 329 patent. She nonetheless refused to allow the patent to be listed against TRIFEXIS because in her view (and in my view as well, as discussed here), the FCA Gilead decision interpreted the product specificity requirement of s 4(2) of the PM(NOC) Regulations as requiring that all the ingredients be listed in the claims themselves.
The FCA has now reversed that decision. Dawson JA’s concurring opinion is straightforward. She was of the view that Gilead is not distinguishable from the case at hand, but that it was wrongly decided and should be reversed [100]. In her view, there is a sufficient nexus between the patent and the subject of the NOC if the patent claims the approved medicinal ingredient, in the sense that the approved product falls within the “boundaries” set out by the claims [107]. In my view Dawson J is entirely correct on all counts.
But the current law is set out in the majority decision, which distinguishes Gilead rather than overruling it. The facts in Gilead (discussed in my posts on the FC and FCA decisions), are also reasonably simple. COMPLERA is formulated with three medicinal ingredients: (1) tenofovir; (2) emtricitabine; and (3) rilpivirine [3]. The 2,512,475 patent at issue includes formulation and compound claims (34 and 48) to tenofovir and emtricitabine in combination with a non-nucleoside reverse transcriptase inhibitor (NNRTI) [28]. Rilpivirine is an NNRTI [10]. It is therefore apparent that a generic version of COMPLERA would necessarily infringe the 475 patent.
In Lilly / TRIFEXIS the majority distinguished Gilead on the basis that the trial judge, Mosley J, held that the patent “did not claim rilpivirine as a medicinal ingredient” [81]. This is a curious use of the word “claim.” I have always understood it “a claim for” something (the phrase used in s 4(2)) as meaning a claim which encompasses that thing. As Dawson JA put it “given the construction of the 475 patent by the Federal Court, and given the agreement of the parties that Complera contained tenofovir, emtricitabine and rilpivirine, and that rilpivirine was a member of the class of NNRTIs, without doubt the 475 patent claimed the combination and formulation of tenofovir, emtricitabine and rilpivirine; any manufacture or sale of that formulation would infringe the 475 patent” [103].
The majority does not dispute any of this. The majority’s view that the 475 patent did not claim rilpivirine was based on two paragraphs in Mosley J’s decision:
I construe the relevant claims of the ‘475 Patent as combinations and formulations of two
medicinal ingredients plus a third one of the NNRTI class that could possibly include but
is not specifically rilpivirine. [79, quoting Gilead FC [26], emphasis added in TRIFEXIS
FCA]
There is nothing in the ‘475 Patent that points specifically to rilpivirine as the third
ingredient in the class of NNRTIs. As the evidence of Dr. Miller on behalf of the
applicant states, several other NNRTI’s had been studied for their efficacy in treating HIV
prior to the grant of the patent. References to an NNRTI in the patent are not to a specific
medicinal ingredient but rather to the class of compounds, one or more of which may
have been found to be suitable to be included in a formulation with tenofovir and
emtricitabine. The claims that specify such a formulation are not specific to the drug in
the Complera NDS. [80, quoting Gilead FC [46]]
Both these paragraphs are to the same effect: while rilpivirine is an NNRTI, it is not the only NNRTI, so it follows that a product that does not include rilpivirine might nonetheless infringe the claims in question. That is certainly true, but it is also true that any product that does include rilpivirine as the NNRTI would necessarily infringe. Thus there is nothing in these paragraphs indicates that 475 patent does not “claim” TRIFEXIS, at least when that word is used as Dawson JA uses it.
It is possible that the majority was using the word “claim” differently from Dawson JA. It might be suggested that the majority is of the view that a patent only “claims” a compound if the compound is specifically listed in the claim – but that cannot be right, because that position is explicitly disavowed by the majority [90] and it is inconsistent on the facts with the court’s unanimous holding in TRIFEXIS. Maybe there is some other meaning of the word “claim” that can explain all this, but I doubt it. In my view, Dawson JA is right: Gilead is indistinguishable from the case at hand [100]. The majority has created an untenable distinction in order to avoid Gilead without expressly over-ruling it.The majority did acknowledge that "certain statements in Gilead FCA arguably led the Judge astray" [93].
It is unfortunate that Dawson JA did not carry the day. The main justification for adhering to precedent is to provide certainty. But an unfounded distinction does the opposite; we now have two irreconcilable decisions that are both formally good law. In principle, this is no different from having no precedent at all, as it is a proposition of logic that any conclusion that be derived from a contradiction.
I suspect that ultimately Dawson JA’s position will prevail. Whatever the correct reading of Gilead might be, the majority has made it very clear that the exact literal words if the ingredients in the product do not need to appear in the claim in order for it to be “a claim for” that product [90]. The majority summarized its position by saying:
the question . . . is not whether the words milbemycin oxime appear in the claims of the
‘329 Patent, but whether the claims of the ‘329 Patent claim milbemycin oxime as a
medicinal ingredient in the formulation set out in the Patent [94].
I suspect that future courts will take this principle from the majority decision, while at the same time using the ordinary meaning of the word “claim” as it was explained by Dawson JA. If that is what happens, the result will be that Dawson JA’s concurrence will effectively state the law, and Gilead will confined entirely to its facts.
However, the majority also said “the patent must claim specifically the formulation which the Minister has approved through the issuance of a NOC” [97, my emphasis]. The requirement that the patent much claim the formulation “specifically” might be understood as something different from merely “claiming” the formulation. In that case Gilead could be distinguished on the basis that while the patent in question claimed COMPLERA, it did not claim it “specifically,” whatever that might mean. I expect, and hope, that future courts will not parse the decision so finely.
No comments:
Post a Comment