TIP Conference Day 3: Pharmaceutical Patents

Posted: February 15th, 2012 | Author: Kevin P. Siu | Filed under: Intellectual Property, Patent, Pharmaceuticals | Tags: TIP Conference | No Comments »

Today was Day 3 of the TIP Conference, which focused on the topic of Pharmaceutical Patents. As today’s panel moderator, Geoff Mowatt from Dimock Stratton noted, the pharmaceutical space is where all the big patent litigation is happening in Canada right now. $22.3 billion is spent annually by Canadians on prescription drugs, of which 58% are patented. The developing case law in this area is therefore very lucrative and high-stakes. This year, about 64% of all Canadian patent litigation will be dedicated to pharmaceuticals alone.

Today’s panel featured three seasoned patent litigators, Don Cameron from Bereskin & Parr LLP, Andrew Shaughnessy from Torys LLP, and Yoon Kang from Smart & Biggar.

Mr. Cameron began by discussing the state of patent law in Canada after the Supreme Court decision in Apotex Inc. v. Sanofi‑Synthelabo Canada Inc., 2008 SCC 61, where Justice Rothstein adopted the four-step test from the UK case law in Windsurfing International Inc. v. Tabur Marine (Great Britain) Ltd. for determining whether a patent is “non-obvious”.

He argued that the elements adopted in the test was flawed, as there are some important differences between UK and Canadian patent law which was overlooked in the decision. For instance, in step 2 of the non-obviousness test, the court is required to “identify the inventive concept of the claim in question, or if that cannot readily be done, construe it.” The requirement for inventive concept is required in the UK Patent Act, but not Canada’s – so “construing and inventive concept” is possibly inapplicable in Canada.

Moreover, British courts have held that experimentation is not “invention” for the purpose of a patent, and therefore results of experimentation alone are not patentable. In Canada, however, this is not the law – although it has been creeping into the jurisprudence. The “obvious to try” test also differs from British law, where courts have held patents invalid if it was obvious that experimentation would work. Thus, it is much more difficult to find a patent valid under the scheme adopted by Justice Rothstein.

Mr. Cameron considered a few more examples of how the test has been applied, and pointed to the case of Abbott Laboratories Ltd v Canada (Health) and Sandoz as an example of where all of the steps had been applied properly. He concluded by summarizing the obviousness test: “a predictable solution may be obvious, but an unpredictable outcome cannot be an obvious solution.”

Mr. Shaughnessy spoke next, discussing the concept of utility and usefulness in patent law. In order to uphold a patent, it must be a “useful” invention – a concept that has been subject to some debate. What is required to satisfy the usefulness criteria?

For starters, the inventor must demonstrate utility at the date of filing – even a “scintilla” of usefulness will suffice. However, there is an underlying tension behind the concept of patents – inventors want to disclose and promise as little as possible, yet courts will always want greater detail in order to uphold the patent bargain – that the inventor discloses her invention in order to get monopoly rights.

One of the greatest difficulties of the utility requirement is that many inventions are not “demonstrably” useful (e.g. through experiments) at the date of filing. When this is the case, the patent can still be valid if utility is based on “sound prediction”.

There are three requirements of sound prediction: (1) a factual basis, (2) an articulable line of reasoning that can link the factual basis to the prediction, and (3) proper disclosure. The third requirement, in particular, raises some tough questions.

This element came from a case about the antiretroviral AZT (Apotex Inc. v. Wellcome Foundation Ltd., [2002] 4 S.C.R. 153), but has, as Mr. Shaughnessy put it, “played like a bad high school prom dance theme” in the courts, being strictly applied in the case law. The proper disclosure requirement for sound prediction means that any data that is needed to predict its utility must be included in the patent application. External data, such as lab notebooks and reports, cannot be used in defending a sound prediction claim.

This in effect sets up a two-tier system, where inventions that can be demonstrated directly by experiments can be defended with external data, while sound prediction claims cannot. This tips the balance of the underlying patent bargain, as it incentivizes demonstrable utility over sound prediction, in effect minimizing what inventors “promise” in their patents. In so doing, however, courts have been reluctant to uphold any patents that do not appear to “promise” anything useful.

The last speaker was Yoon Kang, who further discussed the requirement for disclosure of utility. She noted that this requirement does not find its basis in the Patent Act. This requirement was added by the case law only in 2002 – there is no statutory requirement to show why the invention works or to prove why it works, which is what the proper disclosure requirement does in effect.

Ms. Kang added that in the AZT case, the question of proper disclosure did not actually need to be answered, although it set out this framework which has since been strictly applied by courts.

The disconnect between the two standards – that of demonstrated utility and sound prediction – is hard to explain. It does not come from the enablement requirement in the Act, because any patent being litigated has likely already been implemented therefore it has clearly been put into use. She pointed out that this gives rise to some inherent unfairness, as something that is soundly predicted and provides a real benefit to society might not be patentable because of this proper disclosure requirement.

The situation is exacerbated by the fact that patent applications cannot be modified or added to after it has been filed. This is particularly problematic for many pharmaceutical patents that were filed in the 1990s – before the proper disclosure requirement for sound prediction was added by the Supreme Court. Because those patents were not filed with the new framework in mind, most do not have the required information within the application to support a sound prediction claim. The Supreme Court has in effect set up a moving target for patent law, which has possibly caused millions of dollars of lost profits from patent rights.

 

---