Canadian Supreme Court Voids Pfizer’s Viagra Patent – Focusing Disclosure Requirement on the Patent as a Whole

Teva Canada Ltd. v. Pfizer Canada Inc., 2012 SCC 60 (Supreme Court of Canada 2012)

In a November 8, 2012 decision, the Supreme Court of Canada has ruled that Pfizer’s Viagra patent is void for failure to satisfy the disclosure requirement of § 27 of Canadian patent law. See, Canadian Patent 2,163,446. Apart from the major market impact, the case is doctrinally important in the way that the court treats the disclosure requirement as applying to the patented invention as a whole rather than focusing only on a claim-by-claim analysis.

The active ingredient in Viagra is sildenafil. However, the bulk of the patented claims are more broadly directed toward the use of classes of compounds that include sildenafil or sister compounds to treat erectile dysfunction (ED). Of course, of all of these potential and patented compounds, only sildenafil has been proven effective in treating ED. Further, the patent document does not particularly identify disclose that sildenafil is the compound that works.

The court found that the disclose was lacking because it did not particularly direct one skilled in the art to use the particularly important compound – sildenafil.

Considering the specification as a whole, the use of sildenafil and the other compounds for the treatment of ED comprise one inventive concept. Even though a skilled reader will know that, when a patent contains cascading claims, the useful claim will usually be at the end concerning an individual compound, the claims in the patent ended with two individually claimed compounds [only one of which is sildenafil]. There was no basis for a skilled person to determine which of Claim 6 and Claim 7 contained the useful compound, further testing would have been required to determine which of those two compounds was actually effective in treating ED.

. . .

Although s. 27 does not specify a remedy for insufficient disclosure, the quid pro quo underpinning the Act leads to the conclusion that deeming the patent invalid is the logical consequence of a failure to properly disclose the invention and how it works. If there is no quid — proper disclosure — then there can be no quo — exclusive monopoly rights.

Patent void and invalid.

50 thoughts on “Canadian Supreme Court Voids Pfizer’s Viagra Patent – Focusing Disclosure Requirement on the Patent as a Whole

  1. Not allowed to amend? Paul, that sounds like what today we might regard in Europe as adding matter. Leads me now to think, what provisions of substantive patent law are not involved in this case. By now we’ve covered novelty, obviousness, insufficiency and added matter. That’s all of the grounds under which a European patent directed to eligible matter can be attacked after it issues. But then, how many Opponents to the patent were there at the EPO. If memory serves, 17.

  2. I like your legal articles, especially ones regarding patent law, copyright etc. Patent law issues are very complicated, but you make them absolutely clear. If you are a practicing patent attorney, please submit you contacts to my attorney directory. I invite there only best lawyers and make detailed descriptions about their services and practice areas. For example, look at the category with NY patent attorneys link to attorney-online.info I wish more people to be listed there to provide a good choice for people who need high quality legal services.

  3. Also @ Max

    The relevant quote from the Canadian decision is this:

    “In this case, if we consider the specification as a whole, there is nothing to support the view that the use of sildenafil for the treatment of ED is a separate invention from the use of any of the other claimed compounds for that same purpose. No specific attributes or characteristics are ascribed to sildenafil that would set it apart from the other compounds. Even if we take into consideration the fact that sildenafil is an “especially preferred compound”, there is still nothing that distinguishes it from the other eight “especially preferred compounds”. The use of sildenafil and the other compounds for the treatment of ED comprises one inventive concept.. In fact, the patent itself suggests that the entire class of claimed compounds will be effective in treating ED>”

    If you look back to the 1950′s RPCs, you will find that this was the reason why May & Baker were not allowed to amend down from the only partially active genus that they had disclosed to the two active species that they had exemplified. In your Taxol example, there is no reason why the patentee should be allowed to amend down absent disclosure and supporting data.

  4. I had considered the Taxol case, and there are passages in the Canadian decision which resonate with the facts in Taxol. But I have not yet studied if and how the two decisions can be reconciled. A paper on the whole subject is needed and will be written if time permits.

    The take-away message, though is that Markush claims in non-predictable arts are distinctly suspect and that the valuable claims are the claims to the compounds that have actually been made and tested. The late John Kemp of the firm of J. A. Kemp & Co was aware of this from his involvement in the May & Baker case, and the importance of compound claims was emphasized to trainees joining that firm in the late 1960′s.

  5. Paul it’s the TAXOL coated stent case that was such borderline fun. Suppose:

    1. The app as filed claims coating with ANY anti-angiogenic substance, to inhibit restenosis. Taxol is there as a worked example, but NOT demonstrating any restenosis inhibiting effect, only demonstrating its (already known) anti-angiogenic activity. There is a dependent claim limited to Taxol.

    2. The independent claim lacks novelty so Applicant narrows to taxol which, it even later turns out, is the only member of the class that works.

    As anon reminds us, the claim’s the thing. But what if Taxol was the worked example only because taxol is the only off the shelf readily available anti-angiogenic molecule. In the app as filed, there was no suggestion it was any better a performer against restenosis than any of the other myriad members of its class.

  6. From the preceding arguments, I think that undue deference is given to arbitrarily selected Markush groups.

    What is a patentee in this situation entitled to claim? Firstly the compounds which have been made and tested and for which data shows effectiveness. Secondly compounds for which on the basis of this data there is a rational prediction of effectiveness. As Richard Feynman would have said, these compounds are within the group of things that the inventor has under his control. But NOT any compounds that cannot be predicted to be effective on the basis of existing data. If there is no rational scientific basis for predicting activity, then there is no basis in patent law either.

    Max Drei’s explanation of EPO decisions fit within this framework. To be claimable, the specification should provide sufficient evidence to provide rational basis for the prediction. If there is no such evidence, then the claim scope goes beyond what the inventor has invented and attempts to claim what has not been invented lack merit and should be resisted.

    It should not be forgotten that the evil in the present case is that here the inventors HAD the data but decided NOT to disclose it. They wanted to have their cake and eat it, just like Mr Arkwright in the UK in 1785. There was no test data in the specification that was properly set out so that it could be peer reviewed. As I have said previously this was poor scientific ethics (look at the disclosure obligations in any peer-reviewed scientific journal) and poor patent practice. It is, arguably, inequitable conduct of a very serious kind insofar as patent offices are being deceived that the invention has been disclosed wheres in reality it has not. It is clear that this was at the forefront of the court’s mind when it reached its decision as is apparent from the following passage from the opinion:

    “Pfizer had the information needed to disclose the useful compound and chose not to release it. Even though Pfizer knew that the effective compound was sildenafil at the time it filed the application, it limited its description to the following statement:

    In man, certain especially preferred compounds have been tested orally in both single dose and multiple dose volunteer studies. Moreover, patient studies conducted thus far have confirmed that ONE of the especially preferred compounds induces penile erection in impotent males.

    It chose a method of drafting that failed to clearly set out what the invention was. Even now, in its factum to this Court, Pfizer offers no explanation as to why — knowing that Claim 7 contained the tested and thus, the useful, compound — it elected to withhold that information.”

  7. 1) It’s not necesarily clear: “than all the rest of the originally claimed group” implies that the subset was also an originally claimed group.

    2) It doesn’t really matter all that much, since in the US, we have a no new matter rule.

    3) Whether or not the hypothetical does, my comments about this case still apply (if this case used US law).

  8. Was a separate claim made for the smaller subset?

    According to MD’s hypothetical, I believe the answer is: “the claim did not appear in the application at the time of filing.”

  9. I think MD’s point is that at the time of filing the applicant did not, in fact, “actually think” that the ultimately claimed compounds were “the invention.”

    The claim is the name of the game.

    Was a separate claim made for the smaller subset?

    This yes/no question is not a trick question. Now I have already mentioned that my posts cannot be on Canadian Law as I do not know that law, but applying these facts to US law, wherein claims are separable and can stand on their own even if some of them fail (even fail for close to a Quintilian of reasons), does not make the very specific claims any less valid as “the invention.” The point is that the invention is not the specification or the broadest claim despite a fully enabled and microscopic – but indeed claimed – portion that does meet all of the statutory requirements in its own right. That’s why this decision resonates poorly in the States. The whole thing was tossed when in fact valid claims remained after the invalid ones (should be) tossed.

  10. If the smaller set is what you actually think is your invention, the fact that you are mostly wrong does not change what you are right about.

    I think MD’s point is that at the time of filing the applicant did not, in fact, “actually think” that the ultimately claimed compounds were “the invention.” Rather, in the scenario described by MD, the applicant was merely hoping that that there was an invention lurking in the laundry list but did not have a fixed, concrete idea of what that invention actually was until AFTER the prior art was laid out by the patent office.

    It’s one thing to take a few wild but specific guesses and have one of those guesses turn out to be “inventive.” It’s another thing to simply scorch the earth with a specification that is nothing more than a research plan for testing a laundry list of compositions for their relative “activities.”

    This reminds me, by the way, of one of the more interesting issues that arises with claims to isolated nucleic acid sequences. Although the numbers can get large, there is a finite number of sequences that can be created for any given length of polynucleotide. That means that Applicants can methodically describe and file on millions (or tens of millions, or hundreds of millions) of isolated sequences and poison the well for future applicants while at the same time (thanks to decisions such as the one I alluded to above) preserve their chances at striking gold with a polynucleotide that is later shown to be “surprisingly” useful in a given context. And no, this isn’t just a theoretical strategy.

  11. Not likely. If the smaller set is what you actually think is your invention, the fact that you are mostly wrong does not change what you are right about.

  12. What interests me is the “laundry list” situation where all the compounds are presented in the app as filed as solutions to the technical problem but then, during prosecution, when some of them turn out to be old, evidence is presented which shows that they are not much of a muchness after all but, rather, the selected small group of claimed ones are distinctly better performers than all the rest of the originally claimed group (and therefore patentable over the broader group).

    That selection made after the filing date is a different invention from the one disclosed on the filing date, isn’t it?

  13. Paul: And even if the judges of the CAFC allow patentees to get away with withholding important evidence, in the US a jury in a jury trial may not.

    I”m not talking about “withholding” evidence of non-obviousness.

    I”m talking about filing on laundry lists of compounds for treating X because you want to be first to file and you have not yet collected the necessary data to support non-obviousness, then sometime after filing you “discover” that a third party has published an experiment (after you filed) showing that one of the compounds you listed works surprisingly well. Then you specifically prosecute that compound for treating X and when the USPTO comes back with a obviousness argument you cite the other party’s surprising result. Then when you get your patent, you sue that other party.

    All this is perfectly legal under current CAFC jurisprudence, thanks mainly to one really really ridiculous decision that will surely be overturned but, as always, will fester for a long time.

    a jury in a jury trial

    Not the answer.

  14. Paul Cole mentions the landmark EPO AGREVO decision. I just want to add that it invalidated the claim under the obviousness provision of Article 56 EPC. For me, that’s the way to do it.

    MM reminds us of the possibility of supporting patentability with post-filing experimental data. Yes indeed. Also an everyday routine thing at the EPO. But at the EPO such evidence of patentability isn’t enough by itself. You have to have already made “plausible”, in the app as filed, the invention on which patentability is predicated. What else should one expect, in any “First to File” system of exclusive rights to whole areas of technology, for a term of up to 20 years.

  15. I have not made, and should have made, the observation that at least in the UK and Canada this patent did not stand up in court.

  16. Amongst the 1966 “trilogy” cases before the Supreme Court there was found Adams Battery. In that case the advantages were set out in the specification as filed, and the Supremes observed: “This was not the afterthought of an astute trial lawyer”. Adams was specifically approved in KSR. So there is authority to support the proposition that disclosing evidence for patentability in the specification as filed is helpful.

    The CAFC’s decision that post-filing evidence of surprising results is admissible is not a reason for adopting an everyday practice of omitting important evidence from the specifications that we file. Before the EPO the rule is early and complete disclosure of evidence, and I believe that this is a general rule in virtually all legal systems. And even if the judges of the CAFC allow patentees to get away with withholding important evidence, in the US a jury in a jury trial may not.

  17. Apart from being bad patent practice since if there is data supporting surprising results that is evidence of invention and should be included in the application as filed

    Or should it? Consider that the requirement for disclosure of the best mode is now abolished in the US. On top of that, the CAFC has held (in one of its worst decisions of recent years) that evidence of surprising results can be be provided after the filing date AND can include the work of non-inventor third parties and/or competitors.

  18. Hang on here a sec, Ben. Did you mean:

    “the cost of running efficacy tests on penal tissue samples”

    or

    “the cost of running efficacy tests on penile tissue samples” ??

    The penal ones you can do for the cost of a Playboy, and volunteers will be easy to find. The second one is gonna’ cost you a lot more and volunteers are going to be few and far between.

  19. According to the Canadian court the claims covered a genus of 260 quintillion possible compounds. Of these one worked, and the rest were, apparently, rubbish.

    Pharmaceutical research is based on a link between chemical structure and beneficial biological properties. But in the present case, although there were indications that tests had been performed, these were not disclosed in the patent specification, so that there was no reliable and reproducible data to support the proposition that any of the claimed compounds actually worked. Apart from being bad patent practice since if there is data supporting surprising results that is evidence of invention and should be included in the application as filed, from the standpoint of scientific integrity there has been a clear breach since the data on which the invention is based is not presented for peer review and testing in other laboratories. As Richard Feynman would have said, although there was real science underlying the invention, the disclosure of the patent was no more than cargo-cult science.

    As the Canadian court pointed out, the patentees had possession of the test data but elected not to include it in their specification. Why should they have any benefit for information that they have not disclosed and apparently deliberately decided not to disclose? The strategy of disclosing some information while withholding critical information has not worked since the decision in Rex v Arkwright in 1785, and this is merely a recent example in that unhappy line of cases.

    It is interesting that there was no evidence that any compound within this huge genus had the same biological effects as sildenafil. There are now a number of cases which point to the need to exercise greater care in the drafting of Markush case from May & Baker v Boots (sulfonamides) in the UK in 1950 to the more recent case of Pharmacia/Monsanto v Merck (Cox 2 inhibitors) and the well known EPO appeal Board decision in Triazoles/Agrevo. The sildenafil case provides a further example of a manifestly deficient Markush claim. There is a warning here that should receive attention far beyond Canada.

  20. “so we’re claiming it all”,”

    That would be a rejection on claims being too broad, and therefore NOT a basis for rejection under US 101 jurisprudence.

  21. I think the CDN supreme court used loose langauge in saying this. Because the initial complaint (the Notice of Application)is a request for a prohibition order to get the court to determine if the generic company’s allegations are justified. The generic copany defendant may not even file an “answer” and hence there is no formal prayer for relief. No counterclaims are filed. If the generic company wants some actual relief he is obligated to bring an impeachment action in regular court with a prayer for relief. In this case, as I understand it, no impeachment was concurrently filed. What the Sup. Ct. should have said in its holding is that Teva’s arguments are justified and that the application for the prohibition order is denied.

  22. I see what you mean, but I’m no scholar. we need Paul Cole for that.

    UK law used to offer “lack of support” for the breadth of the claim as a ground of invalidity. But since the EPC came into force in 1978 Europe has had to manage with just three grounds of invalidity a) ineligible/unpaztentable over the art b) insufficient description to enable c) subject matter added during prosecution.

    When that’s all you’ve got, against an over-broad claim, it’s ptretty easy to invalidate on ground b).

    Why have more grounds than a list that is necessary and sufficient? But once you have made out the statutory ground of invalidity, why not add a verbal flourish, to justify the decision to revoke with whatever language best expresses the underlying thinking, be it “covetous claiming” “scope should be commensurate with the contribution to the art” or whatever.

  23. from link to iliplaw.com

    Specifically, here the DC went on to state that the patent was also being revoked because the claims were “covetous”. “Covetous claiming” is a doctrine from English patent law. Over the years I’ve tried to understand the scope of this doctrine, and as best I’ve been able to fathom it (any British practitioners out there who wish to correct me on this if I’ve got it wrong, please do!), it can take one of two forms. In its benign form, it says that a claim (or all your claims) are broader than the specification can support, and are therefore invalid. In other words, “covetous claiming” is just a another way of saying “lack of support” or “lack of enablement”. In its malignant form, this doctrine says that the independent claims are so much broader than the specification can support that it’s clear you should have known you were trying to claim something that’s unpatentable, and therefore as punishment for being so “cheeky” the adjudicator won’t even look at your dependent claims – your whole patent or application is out the window because you included one or more obscenely broad claims.

    Kind of what I said above…appears to be a penalty

  24. to one unspecified member

    That’s the problem with why this is not a red herring.

    The member was specified. It was one of two specifically claimed individual compounds.

    The court decision makes an explicit point that the experimentation was between the two claims (6 and 7).

    On the other hand, rather than “red herring,” it might be a smokescreen, as I mentioned above, this smacks as to a court punishing what it deems bad behavior. The problem is that the court provides a smokescreen (or red herring if you prefer) that does not pass muster.

  25. I think the two-claim issue is red-herring.

    Let me illustrate with a hypothetical containing only one claim. If the patent discloses 1 million compounds, and provide the efficacy/utility data to one unspecified member of the 1 million and no more, is that disclosure sufficient to support a claim directed to the therapeutic use of a single specific compound X?

    What is the experiment we have to envision for ascertaining undue-ness to answer this hypo?

  26. purely – not purity MM.

    Try acting like an adult and leave the insults aside. They just make you look bad when you miss the point (again).

  27. A showing of a need to conduct 578923 experiments is another fact in the world that satisfies the characterization of undue experiment.

    Maybe.

    But maybe not.

    The amount of experimentation is art dependent. 600,000 experiments may be a standard number for a particular art field.

    The difficult item in this decision to understand is how the court thought that experimenting between two claimed items in a starting class of 260 quintillion compounds was at all “undue” by any standard.

  28. I think the conversation between you and MM is exactly what I wonder about.

    Difficulty and unpredictability may be the legal factors. But these are characterizations of facts in the world, not the facts themselves. I understand cost as one evidentiary proof that gives rise to an inference of difficult and unpredictability (not that it is the only one). A showing of a need to conduct 578923 experiments is another fact in the world that satisfies the characterization of undue experiment.

    Going back to my original question, I am still not sure what type of experimentation the court will look to to measure undue-ness. Yes, I know it is whatever a person of ordinary skill does. But I happen to think that’s a cop out.

  29. anon I: Cost has nothing to do with whether or not the testing is undue.

    anon II: Cost alone – and on its own merit – is simply not a factor.

    LOL! Once again we must all bow down before the great Humpty Dumpty. It’s impossible to keep up, really. Where did that goalpost go?

    cost is purely a secondary factor to the primary factor of “difficulty and unpredictability.”

    LOL. And of course whether something is “difficult and unpredictable” has “nothing to do” with whether a particular amount of experimentation is deemed “undue”.

    I can’t wait to find out how one determines the relative “purity” of a “secondary factor” to the “primary factors” of “difficulty and predictability.” Keep the hits coming, anon!

  30. No MM, cost is purely a secondary factor to the primary factor of “difficulty and unpredictability.” Cost alone – and on its own merit – is simply not a factor.

  31. anon:

    Cost has nothing to do with whether or not the testing is undue.

    Not true, at least not in the mind of Judge Newman:

    The district court found that this separation was not a simple or routine procedure and that success in separation, as well as the allocation of properties, was unpredictable…. The court found that Sanofi’s expenditure of tens of millions of dollars for several years of development of the racemate PCR 4099, before separating the enantiomers, also weighed against finding that separation would have been obvious. Again, Apotex has demonstrated no clear error in the extensive finding of the district court concerning the difficulty and unpredictability of the separation of these enantiomers.   These unchallenged findings undermine Apotex’s argument in this appeal that the separation of the enantiomers would have been obvious.  

    SANOFI SYNTHELABO v. APOTEX INC (2008 CAFC)
    link to caselaw.findlaw.com

    The lesson here: if you’re trying to “invent” something obvious, make sure you waste a lot of money on it.

  32. is any testing undue?

    My understanding on this question is that “undue” is related to the art. And that relation is purely “functional.” Cost has nothing to do with whether or not the testing is undue.

  33. This is the score board for this patent. This family is fascinating because they are like Apple’s patents–litigated everywhere with very different outcomes:

    US: One broad claim to cGMP inhibitor invalid for anticipation (Dennis wrote about the Horny Goat)
    UK: Entire patent invalidated due to anticipation.
    Australia: One broad claim to cGMP inhibitor narrowed.
    China: Entire patent invalidated due to insufficient disclosure but later reversed and restored.
    There is also a challenge in South Korea.

    There are two strands of challenges:

    (1) Prior art challenge against cGMP inhibitor claims brought by makers of Levitra and Cialis. These two “me-too” compounds avoid the structure claims but are blocked by the cGMP inhibitor claims.

    (2) Disclosure challenges against sildenafil utility data brought by true generics manufacturers. When China initially invalidated this patent on very similar ground with Canada 10 years ago, the international pressure was brought to bear and China eventually changed its position in favor on its validity to protect Pfizer. Many patent specialists in China consider that episode a political event and believes that the patent should be invalid under the law.

    In summary, this is the problem. The patent is directed to the use of sildenafil to treat ED. The disclosure say that here is the efficacy data for one especially preferred compound (without naming sildenafil), and by the way there are nine of these especially preferred compound.

    I think it is misleading when the court said: “There was no basis for a skilled person to determine which of Claim 6 and Claim 7 contained the useful compound, further testing would have been required to determine which of those two compounds was actually effective in treating ED.” The real problem is not the failure to choose or else, like many said, they can split the application. In any event how can the addition of other claims undermine the validity of one claim?

    A better way to conceptualize the problem is to think hard about whether the claim should issue when there is only one claim directed to the compound sildenafil. Would the efficacy data generated possibly by one of nine compounds and a disclosure that sildenafil is one of the nine compound support a claim directed to sildenafil? What kind of work does a person of ordinary skill need to perform to ascertain that? Or is that the question? You can see this as the old 112/101 enablement/utility pincer attack. Note also that this is a treatment patent, the utility of the invention is the invention. So you have to really show that your claimed compound works as of the filing date instead of using a later affidavit.

    To illustrate the problem in a more dramatic way, consider the case in China which is a true tragicomedy. There, Pfizer claimed only one compound but had a translation hiccup where the “especially preferred individual compounds” where erroneously confused with “particularly preferred compounds” which contained 100+ candidates. So the compound responsible for that data could be one out of 100+ possible compounds instead of one out of 9.

    Here is my question. For undue experimentation, what is the proper experiment? Is the experiment to test each of the 9 (or 100+) compound named in the grouping until one finds the right match for the disclosed utility data that support the claim? Or is the experiment to test each of the claim compound (2 here, 1 in China) until the efficacy is confirmed? Or given the unpredictability and the cost of running efficacy tests on penal tissue samples, is any testing undue?

  34. I’m confused.

    I thought that the validity/invalidity of each claim stood on its own, as if written in independent form. So a dependent claim may be valid, even when independent/base claim(s) are invalid.

    One response to this decision is to file “n” different applications. One for each level of genus and one for each species or even sub-species. Each specification will need to be drafted to the broadness/narrowness of its own claims.

    But is the knowledge disclosed/benefit to the public of these “n” applications greater/better than that made with one application? Or does this decision just cause much more non-valuable work.

    [There is also a possible second issue here: of establishing utility as of the filing date?]

  35. Not sure Ned, as I have not seen the application. You may be entirely correct.

    On the other hand, from the write up here, the distinction is not drawn to a multitude of items not properly described (per se); but rather, that the description is that everything works (including those things claimed that do not work). This appears almost to be a penalty against even those things fully described and meeting the equivalent US 101 and 112 requirements, but also – and here’s the rub – the applicant said the same things about (and claimed) things that did not work and that would fail a utility 101 position. In the US, our notion of each claim stands on its own would prevent an entire patent from annihilation like this one apparently suffers. I do not know Canadian patent law to make any informed guesses as to whether this decision is proper.

  36. since this case deals withe the patentee’s disclosure, and not their use, it is more analogous to a 112 rejection for the third requirement of para. 1. anyone remember that one? Bueller?

    In a pertinent US case, the court said that an allegation of better results from one chemical analog was.not evidence of concealment of the best mode, when routine testing would have revealed whether it was better or not. In re Bundy 209 USPQ 48 (CCPA 1981).

  37. Why can’t a patent holder retain rights to the one that does work?

    There’s a difference between “we discovered this one that works, and we extrapolate that the broad genus also works, so we’re claiming it all”, and “here are a bunch of compounds and a bunch of claims to various numbers of them”.

    And, of course, if you have no idea that any of them work at all, you haven’t actually invented anything.

  38. Thanks Shashank – Interesting point. The court wrote the following "Held: The appeal should be allowed. Patent 2,163,446 is void."  To me that looks like an invalidation. Based upon your information here it seems is that this particular holding would not control the outcome of a later infringement action. Right? 

  39. Just to be clear, the CDN Supreme Court did not HOLD the patent invalid. The nature of the proceedings were PMNOC, which are adminstrative proceedings about getting generic drug approval. Under the PMNOC regime, there is no actual holding of patent noninfringement or invalidity; just that if the case of noninfringement or invalidity is strong enough, then the court will not order Health Canada to deny drug approval. This is why, under Canada’s double jeopardy law, if the Brand company “loses” the PMNOC lawsuit, it can still file another lawsuit in “real court” and relitigate the whole case again, esp. when the generic company launches after PMNOC approval. In the levofloxacin case, after having “won” at PMNOC levels, Teva launched, Janssen re-sued on the same patent and claim and won.

    So here, the CDN Supreme Court has not technically invalidated the patent (providing automatic freedom to operate), rather has now allowed Health Canada to issue the NOC approval. Pfizer can still sue for regular infringement if it wants. Though, the trial court will be heavily influenced by the CDN Supreme Court’s pronouncement on the patent invalid.

  40. So this means that Canadian applicants can’t just recite a laundry list or broad class of compositions in the context of a treatment for X and expect to obtain a valid patent on the one that actually works. How unfair.

    the patent document does not particularly identify disclose that sildenafil is the compound that works

    It’s impossible to imagine how such a thing could happen.

  41. This seems to be a truly absurd decision. The fact that Pfizer invented TWO compounds invalidates the claim for the one that is approved? Really?

  42. definitely not in accord with US 101 jurisprudence.

    It’s apparently not in accord with Canadian jurisprudence either, since the statute doesn’t provide a remedy (bonus question – does the public therefore have no right to a sufficient disclosure?), but sometimes a Supreme Court will use its collective brain and come to the right decision anyway.

  43. …. Yes perhaps for the claim that does not work, but to throw out the claim that does work – in your analogy, that does have utility – is definitely not in accord with US 101 jurisprudence.

  44. Makes sense to me.
    If the applicant has discovered that there is ONE specie that works and the applicant also knows that the other species do not work, then any claims covering an invention beyond that one specie should be invalid.
    This seems equivalent to a US rejection under §101 for lacking utility.

Comments are closed.