Selection patents recently received a huge vote of confidence from the Supreme Court of Canada (SCC). In a decision rendered on November 6th, 20081, the SCC rejected Apotex’s attempts to both invalidate Sanofi’s selection patent (‘777), which covers the anti-coagulating drug Plavix, and have the SCC render selection patents invalid in Canada altogether.  

Sanofi’s earlier ‘875 patent disclosed a genus which covered more than 250,000 possible compounds useful for inhibiting platelet aggregation. One compound, a racemic mixture of two enantiomers, dextro-rotatory clopidriogel and levorotatory clopidriogel, was of particular interest to Sanofi. Sanofi discovered to their surprise that the dextro-rotatory enantiomer had a higher therapeutic effect and lower toxicity when compared to the levo-rotatory enantiomer and the racemic mixture, and were granted the ‘777 patent based on this discovery.

Pharmaceutical Selection Patents

The pharmaceutical industry uses selection patents to protect inventions when they discover a surprising and unexpected advantage of a selected compound over a genus of compounds from which the selected compound was chosen. The SCC referred to the judgment in In re I. G. Farbenindustrie A.G’s Patents (1930)2 where it is established that if the selected compound is new and if it has properties of an unforeseen nature, then the inventive character is met. The SCC confirms that a selection patent does not differ in its nature from any other patent and that it is consequently compatible with the law on the patents.  

Apotex alleged invalidity of the ‘777 patent on prior art grounds, namely that Sanofi’s earlier ‘875 patent anticipated the ‘777 patent and that the ‘777 patent was obvious; and on double-patenting grounds.


To address the issue of anticipation, the SCC adopted a two-step approach to establish whether a prior publication is indeed an anticipatory piece of prior art. The approach requires separate consideration of “prior disclosure” and “enablement”, both of which if proven, establishes that the prior art is anticipatory. In the context of selection patents, if a person skilled in the art, when reading the genus claim in the prior patent, does not discover the special advantages of the selection, then there is no anticipation by way of “prior disclosure”. No trial and error is allowed. Similarly, if the person skilled in the art has to perform the invention with undue burden, then there is no anticipation by way of “enablement”. A skilled reader of the patent should be able to arrive at the invention the first time they try it and each time after, with limited trial and error. In Sanofi’s case, there was no anticipation because their former patent does not reveal which enantiomer is more active even if it is known that an enantiomer is often more active than another. Since anticipation requires both prior disclosure and enablement, it was unnecessary for the SCC to consider the second issue of enablement. Even if the methods of separation of enantiomers were known, the fact is the Sanofi inventors had to spend several months of research to identify the active enantiomer and its advantages. Therefore, the ‘875 patent was found to not anticipate the ‘777 patent as alleged by Apotex.  

Interestingly, this case provides some guidance to patent practitioners who draft selection patent applications and who are often confronted with the question of how many comparative examples does one need in the application. Sanofi included twentyone comparative examples out of a possible 250,000 and clearly articulated the special advantages of the selected dextro-rotatory enantiomer in the ‘777 patent. The SCC found this acceptable. Compare this number to a case before the Federal Court earlier this year, where GlaxoSmithKline’s selection patent3 for an orally bioavailable ester of acyclovir was found invalid for lack of utility based inter alia on only three comparative examples being presented in the selection patent. Clearly, for a valid selection patent the number of examples will depend upon the nature of the invention and on the number of compounds found in the original genus claim. It is safe to assume that a number that is a sufficient representative of testing is required to establish a valid selection, especially in the pharmaceutical arts- “the more the merrier” seems to be the order of the day.  


Historically, the Courts have relied on the rather strict language of Beloit4 to decide obviousness cases and whether the notional skilled person, in light of the state of the art and common general knowledge, would have been led directly and without difficulty to the solution taught by the patent. Also the Courts have routinely rejected an “obvious to try” criterion. Interestingly, the SCC compared current Canadian attitudes towards the question of obviousness with those of the United Kingdom and the United States, where a stringent, but not mandatory, “obvious to try” approach is the current state of the law. The SCC examined the circumstances under which it might be appropriate to use the “obvious to try” criterion and concluded that for an invention to be obvious instead of it being merely “obvious to try” it must also be “more or less self-evident to try to obtain the invention. Mere possibility that something might turn up is not enough.”

When considering obviousness from an “obvious to try” perspective, the SCC took the lead from Windsurfing Internationa5l where the question “[when] viewed without any knowledge of the alleged invention as claimed, do those differences constituted steps which would have been obvious to the person skilled in the art or do they require any degree of invention?” was posed. In each case, where the Courts are confronted with the question of whether an invention is “obvious to try”, the evidence will be considered with the following questions:  

1. Is it more or less self evident that what is being tried ought to work? Are there a finite number of identified predictable solutions known to persons skilled in the art?  

2. What is the extent, nature and amount of effort required to achieve the invention? Are routine trials carried out or is the experimentation prolonged and arduous, such that the trials would not be considered routine?  

3. Is there a motive provided in the prior art to find the solution the patent addresses?  

The SCC found that Apotex failed to establish that the selected dextro-rotatory isomer was “obvious to try” from the 250,000 other possibilities in the ‘875 patent. If it were, Sanofi would hardly have spent the time and resources to develop clopidriogel bisulphate. For the “obvious to try” test to work, the invention should be more or less self-evident to the person skilled in the art. Therefore, the SCC found the ‘777 patent valid despite Apotex’s allegation that it was obvious.

Double Patenting

Canada’s Patent Act provides for “one invention per patent”. Apotex argued that selection patents extend the lifetime of the original genus patent by allowing so-called “evergreening” or double-patenting. The SCC rejected this argument, pointing out that selection patents can be obtained by third parties and that they encourage innovation by identifying hitherto unknown and useful properties in the original genus. In a double-patenting challenge, the focus is on the claims of the two patents rather than on the disclosure and because the claims of the genus patent are broader than those of the selection patent, there cannot be double patenting. Therefore, the SCC concluded that the ‘875 patent was guilty of “same invention” double patenting in light of the ‘777 patent as alleged by Apotex.  


The SCC now appeared to recognize the need to soften the stringent test for obviousness in Beloit. Although the SCC’s decision seems to bring the test closer to that applied in US and UK courts, the Canadian test for obviousness test is still more rigid than the US and UK. It remains to be seen, however, whether Sanofi’s victory will pave the way for more selection patents in Canada and what effect this will have on the way selection patent applications are examined at the Canadian patent office. In the meantime, we can expect that selection patents will be scrutinized for compliance with the aforesaid criteria and that the Courts have not heard the last of this.