Canadian Supreme Court Confirms Plavix Patent Despite Evergreening Charges

Apotex Inc. v. Sanofi-Synthelabo Canada Inc., 2008 SCC 61 (Canada Supreme Court 2008)

The Supreme Court of Canada has upheld Sanofi’s “selection patent” (the ‘875 patent) covering the blockbuster anti-clotting drug Plavix. A broader “genus” patent (the ‘777 patent) – also held by Sanofi – discloses over 250,000 different compounds. The narrower patent does have some additional patent scope. In particular, while ‘875 patent claims a particular dextro-rotatory isomer, the prior ‘777 patent only describes a racemate of that compound that includes two different isomers.

As an apparent “evergreening” strategy, Sanofi did not claim priority to its earlier patent filing. This allows for an additional decade of patent term. Looking at the case, the Supreme Court determined that it should simply follow the law of novelty and obviousness and not create any special ‘evergreening’ exception:

[97] Evergreening is a legitimate concern and, depending on the circumstances, strategies that attempt to extend the time limit of exclusivity of a patent may be contrary to the objectives of the Patent Act. The Act aims to promote inventiveness by conferring exclusivity for a limited period of time while providing for public disclosure of the invention to enable others to make or use it after expiry of the period of exclusivity.

[98] However, a generalized concern about evergreening is not a justification for an attack on the doctrine of selection patents for two reasons. First, a selection patent may be sought by a party other than the inventor or owner of the original genus patent. In such a case, anticipation or obviousness may be an issue, but evergreening does not arise.

On obviousness, the court found “a significant difference” between the claimed patent and the prior genus patent, and that the “difference was not obvious.”

17 thoughts on “Canadian Supreme Court Confirms Plavix Patent Despite Evergreening Charges

  1. 17

    Ummm, evergreening is such a minor part of the decision. The wider impact of this case is “obvious to try” may be an inquiry to determine patentability. However, possibly can’t be used in all arts. Anticipation now includes a second step of enablement. Now in Canada a patent application needs to have a “sufficient description” to be useful. However, if it is to be anticipated the prior art must be “enabled”.

  2. 16

    Norman-

    Given your good points, this appears to be a reasonable legal decision based on bad “facts.”

    Regarding whether this case would come out in the same way in the U.S., I don’t think it’s perfectly clear. Given a mixture of stereoisomers, I think a U.S. court could find that a PHOSITA would expect one of them to be responsible for a lion’s share of the activity, making the isolation of that isomer an obvious advance.

    The more persuasive fact (in my opinion) appears to be the toxicity data. If the racemate had toxicity problems, I don’t think a PHOSITA would expect resolving a single stereoisomer to alleviate these problems.

    Thanks for your insightful comments.

  3. 15

    Nice Llama

    Re obviousness, you state: “the PHOSITA certainly knew of the good activity of the mixture and must have expected that all or most of that activity was concentrated in one stereoisomer.” I don’t doubt that this is true today and it may even have been true at the priority date, but the SCC decision was expressly premised on the finding by the applications judge that the applicant had NOT established that at the priority date the PHOSITA would have expected that the activity was concentrated in one stereoisomer. I don’t know enough chemistry to have my own opinion as to whether this conclusion of fact is plausible, but if we accept it, my understanding is that the SCC conclusion of obviousness would follow even in the US.

    Note that this decision does not imply that isomers are per se patentable over the racemate in Canada. This originated as the Canadian equivalent of an ANDA and I’m not sure how much litigation effort went into the original application. I wouldn’t even be surprised if Apotex went ahead despite this factual finding on obviousness because if it did manage to win nonetheless, it would have been able to attack many selection patents on a purely legal basis, without the need to establish obviousness on the facts (thought that is purely my speculation).

  4. 14

    Norman–

    Regarding, the U.S. claims to the specific isomer of at issue in this case, you write “the answer to MaxDrei’s question is that in the US you would get the molecule per se.”

    While correct, I believe that this answer is a bit oversimplified. We should begin with claim 1 of the U.S. selection patent, which recites:

    “Dextro-rotatory isomer of methyl alpha-5(4,5,6,7-tetrahydro(3,2-c)thieno pyridyl) (2-chlorophenyl)-acetate substantially separated from the levo-rotatory isomer and its pharmaceutically acceptable salts.”

    This claim is certainly a compound claim, covering the SEPARATED “molecule per se.” The claim is far more limited than a typical chemcial compound claim. Specifically, the “substantially separated…” language excises stereoisomeric mixtures from the scope of the claim. Thus, unlike a conventional compound claim, the claim recited above is to only substantially separated (i.e., absent the L-isomer) forms of the specific stereoisomer. To the contrary, a conventional compound claim, such as “A compound of formula A…” would read on a compound of formula A and any compositions/mixtures comprising it.

    For the practitioner this raises an interesting opportunity. Applicants in this application drafted the claim as “…substantially separated from the levo-rotatory isomer…,” which limits the claim to being substantially free of the L-isomer. Alternative drafting could have claimed a “non-racemic” form, which would only limit the claim to non-50-50 mixtures of the stereoisomers, thereby reading on mixtures still comprising the L-isomer albeit in different proportions. This could be practically important where separation techniques are less-than-perfect or the isolated “pure” compound degrades through some type of process that scrambles the stereochemistry of a previously separated form.

  5. 13

    Norman–

    I agree with your position that “[o]bvious to try,” like many other attempts to describe the contribution of an ordinary technician may provide evidence of obviousness but it not necessarily determinative.

    However, your comment that “…it was not obvious at the time [that the application on the isolated stereoisomer was filed] that the properties of the isomer would be different from that of the racemate,” is not the correct analysis in the U.S.. The correct analysis should focus on whether the PHOSITA (from step three of the Graham analysis) would find it obvious to isolate the pure compound given the teachings in the prior art (steps 1 and 2 of the Graham analysis). As a chemist, I believe that isolating a stereoisomer from a mixture of stereoisomers is almost certain to provide a compound of higher purity compared to the mixture. Here, the PHOSITA certainly knew of the good activity of the mixture and must have expected that all or most of that activity was concentrated in one stereoisomer. This is just a consequence of our living in a chiral world.

    Your focus on the “properties of the isomer” is most relevant to initially determining the obviousness of the isomer. These properties provide evidence (aka “secondary considerations”) of non-obviousness which may elucidate an otherwise ambiguous case. This is not a close case. Here, an unmeasured difference in the properties of stereoisomers, where a PHOSITA would expect some difference to exist, should not make isolating a single stereoisomer nonobvious. This isolation is a the contribution of a technician not an inventor. Of course, the real magician here is the patent attorney.

    While I also agree that “improvements are not necessarily obvious over the original invention,” I must also point out that they are not necessarily nonobvious either. Here, for the reasons above, isolating a compound from a mixture comprising it is not the sort of technical advance that our Patent system seeks to encourage. This advance would occur absent the incentive of an exclusive monopoly; granting an exclusive monopoly here would impose societal (economic) costs that far outweigh the reciprocal benefit.

  6. 12

    @Federally Circuitous:

    Canada has been a 20-years-from filing regime since October 1, 1989. The genus/species patents at the heart of this case were filed before 1989, so their term is 17-years-from grant.

  7. 11

    Federally Circuitous — “Finally, as someone who does work in the pharma field, this post was of mild interest, but I don’t practice in Canada – that’s what I have Canadian associates for… academic gobbledy-gook, and I’ll be really disappointed if it further loses its focus by becoming a news sources for many other jurisdictions.”

    Drugs flow from Canada into the US. People from around the world read this blog. Denis is an academic.

    Anything else I can help you with?

  8. 9

    Federally Circuitous — “So I don’t understand the basis for DC’s underlying assumption that a continuation would have been available here.”

    The basis is S.36 of the Canadian Patent Act, which allows for divisional applications.

  9. 8

    The answer to MaxDrei’s question is that in the US you would get the molecule per se: see US Pat 4,847,265 which corresponds to the Cdn ‘875 selection patent; and US Pat 4,529,596 corresponding to the Cdn ‘777 genus patent. If this is “evergreening” (which it is not, as other commenters have pointed out), it is allowed in the US as well.

    There is nothing legally remarkable going on in this case. Apotex’s best argument is that it was obvious to try to isolate the isomer given that it was known that the orignal compound was a racemate. While “obvious to try” may be a relevant inquiry, is not the standard for obviousness anywhere; and in any event, on the facts, it was not obvious at the time that the properties of the isomer would be different from that of the racemate. Apotex also argued that the genus patent necessarily anticipates the selection patents. Answer: no, improvements are not necessarily obvious over the original invention (duh). The evergreening argument was that a selection patent is necessarily invalid for double patenting. The SCC explained that no, a patent for an improvement is not necessarily invalid for either “same invention” or “obviousness” double patenting. In context the evergreening excerpt given above was saying “if there is such a thing as evergreening, it would be a concern – but this isn’t it.” Apotex was raising evergreening as a scare tactic. The selection patent wasn’t evergreening and the scare tactic didn’t work. The only puzzle is why leave (cert) was granted.

  10. 7

    A thread on the patentability of “evergreening” selections, in the USA, would be interesting for me. What I mean is when the art discloses a racemic mixture and then an inventor files for one component of the racemate, with a clever method for its isolation and data revealing its efficacy. In the UK that would be the Lundbeck case. But, in the USA, would you get the molecule per se accepted as novel and non-obvious, or just the clever method for its isolation? Apologies in advance, if this is all old hat to other readers.

  11. 6

    I feel like we’re missing some facts here. Correct me if I’m wrong, but Canada is a 20-years-from filing regime, not a 17-years-from grant regime like the US used to be. So the patent in question was filed 10 years after the earlier patent. How then could Sanofi have claimed Paris priority from the earlier application – the later app was filed well after the priority year was over? Also, as I recall, Canada (like the rest of the world) doesn’t have continuation/CIPs like the USA. So I don’t understand the basis for DC’s underlying assumption that a continuation would have been available here.

    Finally, as someone who does work in the pharma field, this post was of mild interest, but I don’t practice in Canada – that’s what I have Canadian associates for. What’s next, an update on patent practice in Madagascar? I miss the days when this blog concentrated on CAFC decisions, without the academic gobbledy-gook, and I’ll be really disappointed if it further loses its focus by becoming a news sources for many other jurisdictions.

  12. 5

    Please read the above comment as:
    Selection inventions [at least in pharma space] have been recently held valid in UK too – notably the Olanzapine case.
    [2008] EWHC 2345 (Pat)

  13. 4

    Selection inventions [at least in pharma space] have been allowed earlier too in Canada – notably the Olanzapine case.

  14. 3

    Sometimes this blog is here for DC to edify us …

    sometimes it’s here for us to keep DC in touch with practice …

  15. 2

    What would be the purpose of making a priority claim when the new invention is novel and non obvious over the prior application?

    Wouldn’t the priority claim would be ineffective for claims reciting new matter which in this case is apparently all of them?

Comments are closed.