Is It Easier to Invalidate a Patent in Canada? Eli Lilly Thinks So and Wants It to Stop
The Canadian generic pharmaceutical industry has recently been successful in invalidating several brand name drug patents on the basis of “The Promise” doctrine. Eli Lilly would like to put an end to this, using Canada’s international treaty obligations under NAFTA.
Can you see the difference?
In the mid 2000s one could start to see Canadian patent cases “turning” somewhat. Before this, the general sense was that a mere scintilla of utility was enough to obtain a patent. However, if the patentee made an explicit and unequivocal “Promise” of a certain use or result, recent cases have held the patentee to this result. Eli Lilly’s selection patent for an antipsychotic agent (olanzapine) was first held invalid in 2007 (in preliminary type proceedings) (2007 FC 596). Eli Lilly’s patent promised that its compound was better than the rest. However Eli Lilly had not actually determined its Promise, nor was their Promise soundly predictable (ie. it was a guess). In 2011 the generics were again successful in invalidating this same patent. (Eli Lilly v Novopharm 2011 FC 1288; aff’d by bench 2012 FCA 232, appeal pending to SCC). Lilly characterizes this finding as “absurd” in its November 7, 2012 NAFTA Notice.
Interestingly the UK Court of Appeal made a particular point of saying it would not follow the Canadian “Promise” approach in considering the same Eli Lilly patent. The UK court stated that counsel
invited us particularly to follow the Canadian decisions (where the patent was held invalid) …. [the Canadian court] essentially followed the IG approach in saying that the Patent had to disclose an advantage over the prior art. It does not seem he was shown the EPO case law (why should he have been?). It does not seem that he was invited to consider patentability on the basis of whether the Patent disclosed real advantages over what was actually known. So I decline [counsel’s] … suggestion.
(Dr. Reddy’s Laboratories (UK) Limited v. Eli Lilly and Company Limited, [2009] EWCA Civ 1362)
It was clear this was coming
The Supreme Court of Canada (“SCC”) recently invalidated Pfizer’s Viagra patent based on insufficiency (Teva v. Pfizer 2012 SCC 60 (“Teva 2012”)). The “Promise” question was not a live issue for the SCC. Pfizer complained that the “enhanced disclosure requirement” for sound prediction (ie. what is in the patent to back up “The Promise”) violates Canada’s international treaty obligations (para. 88). The SCC simply stated that even if an enhanced disclosure requirement exists, and violates any treaty, this was not in issue in the particular case before it. The SCC in Teva 2012 was only concerned with actual utility (ie. there was no guessing).
Query whether the SCC would like to clarify the disclosure aspects of sound prediction that this same court established 10 years ago in “AZT”.
Nothing in this passage [from AZT] suggests that utility is a disclosure requirement; all it says is that the utility required for patentability … must, as of the priority date, either be demonstrated or be a sound prediction.
The SCC will likely evaluate the disclosure requirements of sound prediction again another day.
Eli Lilly’s NAFTA challenge
Eli Lilly’s complaint is based on an expropriation (ie. theft) of their atomoxetine patent because it was invalidated in Canada based on a violation of international law. NAFTA and TRIPS are treaties that both refer to the mandatory obligation to grant patents for inventions that are “capable of industrial application” or “useful”. Eli Lilly says Canada has failed to enforce patents that meet this requirements.
According to Lilly, Canada now requires regulatory-type testing before a patent can be filed – which obviously has a chilling effect/is not very practical, and discriminates in the area of therapeutic/pharmacologic patents. Lilly’s challenge is based on its patent case regarding the ADHD drug atomoxetine (2010 FC 915, 2011 FCA 220). Even actual clinical trials (because they were too preliminary at the time Lilly filed its patent) were not enough to meet The Promise.
Lilly contrasts Canada’s requirements with other jurisdictions:
– it is acceptable to “guess” in a patent in the US: “The law presumes that if there is an asserted [real world] utility that is ‘credible’, rejection for want of utility is inappropriate”, and it is “rare” to challenge credibility. “To deny a patent on an invention requires proof of inoperability or inutility.”
– Europe also has a low standard – as a use must only “be specified and ‘plausible’, a standard that is met unless …[it ] is…’merely speculation’ and therefore implausible.”
In both US and Europe, if a use is doubtful/inherently implausible, patentees can augment their patents by providing evidence of actual utility (dated well after they applied for their patent).
Eli Lilly fails to mention that the UK case it cites in fact accepts that a guess is enough – it uses the following thresholds interchangeably: “plausible”/ “reasonably credible”/ an “educated guess”/ a mere “possibility of exploitation”.
Is there a real difference? This all seems to be a matter of degree, and context. The devil may very well be in the details – if a use is “plausible” or possible, is it necessarily also soundly predictable – or is it still a guess?
Eli Lilly also says Canada cannot impose “higher” disclosure requirements on it – in particular Canada cannot compel patentees to include evidence of utility in their patents. Lilly fails to acknowledge that there may be good reason for a patentee to decide to include a specific Promise in their patent. For example, Lilly’s Notice ignores the fact that its olanzapine patent was a selection (from a known general class of compounds), and that Lilly had specifically stated in its patent (to back up its selection) that olanzapine has “marked superiority and a better side effects profile than prior known antipsychotic agents, and…a highly advantageous activity level.
Lilly also complains it is in a “Catch 22” situation – it needs to include big advantages/promises in its patent (to get around obviousness objections), but then complains that it must be held to such promises. From a common sense perspective, the public would expect that if Lilly has based their new patent on new advantages over the prior art, then they should be held to demonstrating or predicting such advantages.
Lilly objects that the “Promise” doctrine is being improperly applied only to pharmaceutical patents. Lilly ignores 2 points. Firstly this approach has been applied to non-pharmaceutical cases as well – for example most recently with helicopter landing gear (Eurocopter v. Bell Helicopter Textron Canada Limitée (2012 FC 113)). Secondly, there may very well be practical concerns at play that mean pharmaceutical patents are particularly susceptible to the Promise doctrine. British courts have repeatedly referred to the “special” problems that may arise in the area of pharmaceutical and biologic patents. Unlike inventions for wafer-boards (the invention in a 1981 SCC Consolboard case that is always cited by courts considering utility/sufficiency), pharmaceutical inventions may be in danger of being merely guesses, or not inherently plausible. Guesses should be unacceptable from a public policy perspective – irrespective of whatever “legal” term you consider applicable.
What next?
It appears Eli Lilly seeks to: abolish any substantive utility requirement; and permit a patentee to show utility based on tests done a decade after they file their patent. In addition, Lilly says it should be enough that Lilly had a “secret” sound prediction – ie. the public has no entitlement to see any basis for such a prediction in Lilly’s patent.
Lilly states that it seeks to
[1] rectify the situation of non-compliance resulting from the Judge-made law on utility and disclosure and [2] remedy the violations of the investment obligations owed to Eli Lilly.
Lilly threatens arbitration and $100 million damages.
Eli Lilly may submit a formal request for Arbitration in February 2013, should discussions between Lilly and the Canadian government fail. Such arbitration may only assist with a monetary award. According to the government’s website – should the matter go to arbitration, the tribunal has only limited rights. If there is a violation of “a specific obligation” under NAFTA, “a tribunal can award damages to an investor but cannot … require a NAFTA Party to amend the measure that gave rise to the dispute” (ie. as would be expected, the tribunal cannot amend the Patent Act, or direct any “correction” of jurisprudence).
Query whether other brand names (most notably Sanofi after their loss of their antihypertensive ramipril patent in 2011) will join this fight (or perhaps they already have privately in the “background”).
It is too bad that this piece is entirely written without asking anyone expert in Canadian patent law whether Lilly’s claims are accurate. In fact, they are not. The recent SCC decision in Teva v. Pfizer specifically (and contrary to what was asserted in the article) stated there was NO higher disclosure requirement for sound prediction. In any case, sound prediction and promise of the patent are very different concepts. What Lilly is really complaining about is that it and its legal team did a poor job of arguing before the courts. Rather than take responsibility for this, they blame Canadian jurisprudence which, taken as a whole, is fully in line with international standards.
This article has touched on a number of complex issues that are tied in with the promise doctrine. Among them is that the promise doctrine goes hand in hand with claims construction.
The promise doctrine was initially only envisioned for use where an express promise regarding utility has been made within the patent. Query, however, whether ‘promises’ made in the disclosure, as opposed to the claims of the patent, which ultimately delineate the boundaries of the patent, should be grounds for invalidating the whole patent. Put another way, how can one reconcile the claims construction analysis with the promise doctrine when it comes to determining the utility of the patent?
Further, let us not forget that the SCC has on multiple occasions reaffirmed the principle that a mere scinitilla of utility is needed for utility to be demonstrated, which is a fundamental aspect of Canadian as well as European and U.S patent law.
In the end, regardless of whether one is acting from perspective of a generic or innovator manufacturer, the Courts must better lay out a test for determining what constitutes whether a promise has been made, giving clarity to this rather fuzzy and inconsistent doctrine. Moving towards consistency with the EU and the USA, however, when it comes to utility, would be the most preferable option from my perspective.
I’d also like to point out that in the first paragraph of this article, in the sentence, “However Eli Lilly had not actually determined its Promise, nor was their Promise soundly predictable (ie. it was a guess).”, determined should most likely read demonstrated.
Contrary to Richard Gold’s assertion, the Canadian promise of the patent doctrine, the requirement to disclose the factual basis for sound prediction in the patent, and the prohibition on using post-filing evidence to establish utility, are not in line with US, UK and EPO jurisprudence, at least: see my articles Must the Factual Basis for Sound Prediction Be Disclosed in the Patent?, (2012) 28 CIPR 39 and The False Doctrine of False Promise, 29(1) CIPR (forthcoming). Whether Canadian law on these points is justifiable is a different question. The SCC in Teva v Pfizer did NOT state that there was no higher disclosure requirement for sound prediction. What it said, at para 43, is “Since sound prediction is not an issue, the question whether there is an “enhanced” or “heightened” disclosure requirement with respect to sound predictions does not arise in this case and need not be addressed.”