Case: Apotex Inc. v. sanofi-aventis (T-644-09); sanofi-aventis and Bristol Myers Squibb Sanofi Pharmaceuticals Holdings Partnership (T-933-09)

Drug: PLAVIX® (clopidogrel bisulfate).

Nature of case: Patent impeachment and Infringement actions

Successful party: Apotex Inc.

Date of decision: December 6, 2011 (confidential reasons); January 3, 2012 (non-confidential reasons)


On December 6, 2011, the Federal Court (per Boivin J.) held in a 247-page decision that sanofi-aventis’ patent for clopidogrel bisulfate (PLAVIX®) is invalid because the inventors failed to disclose the factual basis and line of reasoning for their sound prediction of utility in the specification.

The ‘777 Patent

Canadian Patent No. 1,338,777 is a selection patent. Clopidogrel, the compound claimed in the ‘777 Patent, is the dextro-rotatory enantiomer of a racemate that was broadly covered by an earlier sanofi-aventis patent (the ‘875 Patent), along with 9.5 million other compounds.

The invention described in the ‘777 Patent “relates to the dextra-rotatory enantiomer of [PCR 4099], a process for its preparation and pharmaceutical compositions containing it”, as well as specific salts of clopidogrel. It differs from the prior art because “in an unexpected manner, only the dextro-rotatory enantiomer exhibits a platelet aggregation inhibiting activity”. In addition, “the salts of the dextro-rotatory isomer have a better therapeutic index than the salt of the racemic mixture”, whereas the “levo-rotatory isomer exhibits almost no platelet aggregation inhibiting activity and its toxicity is markedly higher than that of its dextro-rotatory homologue.”

The ‘777 Patent describes various tests on clopidogrel, including an in vitro platelet aggregation assay and in vivo metabolism, thrombosis and toxicology studies in rats. Routes of administration for “the medicine” (viz. clopidogrel) and pharmaceutical formulations are also described. The patent then states that “on account of its interesting inhibitory properties towards platelet aggregation and its interference in the mechanism of the formation of arterial and venous thromboses, the medicine of the invention can usefully be administered in the treatment and prevention of platelet disorders due to extracoporeal blood circuits or the consequence of complications in atheroma.” Boivin J. held that clopidogrel is the “medicine of the invention” described in the specification.

The ‘777 Patent ends with 11 claims, including: clopidogrel and its salts; a process for making clopidogrel; and pharmaceutical compositions of clopidogrel. The claims do not specify any particular use of clopidogrel. However, the Court held that the ‘777 Patent “relates to one invention and that can be described as a compound which is useful in inhibiting platelet aggregation, has greater therapeutic effect and less toxicity than the other compounds of the ‘875 Patent, has the advantages of the salts (crystallize easily, not hygroscopic and sufficiently water-soluble) and the methods for obtaining that compound.”

Promise of the Patent

Apotex argued that the ‘777 Patent “guarantees” that the clopidogrel is useful for treatment in humans. By contrast, sanofi-aventis argued that the patent only promises a “potential” use in humans.  Boivin J. was “unable to accept either of the two extreme positions”. Instead, the ‘777 Patent was interpreted as promising that clopidogrel “can be used in the treatment of certain human thrombotic diseases”. Even though there was no reference to humans or medicinal uses in the claims, Boivin J. held that the “wording used throughout the patent”, including medicine, patient, pharmaceutical compositions, active pharmaceutical index and “medicine of the invention”, as well as the list of “human diseases” on page 21 of the specification, is consistent with an explicit promise of human utility. In the Court’s view, “matters of medicine in the ‘777 Patent are inescapable”.

The Court held that in order to assess the promise of a selection patent, it is “helpful” to consider the teaching of the genus patent from which it was selected.  Boivin J. noted that in the '875 Patent, the inventors stated that the claimed compounds are “very useful in human and veterinary therapeutic applications”. Boivin J. held that “it would be illogical to allow a selection patent to promise less than a genus patent” in the present situation.

The Court held that events giving rise to the development of the patented compounds could also be useful when evaluating the promise of the patent. Boivin J. rejected sanofi-aventis’ “matching principle” argument and the notion that the promise of the patent must be construed in light of the information contained in the patent and how it would be interpreted by a person skilled in the art. The Court held that this would lead to the “illogical result that the promise of the patent could never be in humans if there is no human data in the patent itself.” 

Utility and Sound Prediction

The Court held that the promised utility of the claimed compounds had not been demonstrated at the filing date. Boivin J. noted that it in analyzing the utility of a selection patent, the Court should not consider each specific advantage in the disclosure separately. Rather, utility must be considered in conjunction with the invention as a whole. Although the evidence established that a double-blind Phase I trial was underway at the filing date, Boivin J. held that it was unclear whether the results were available to the patentee. In any event, Boivin J. held that the results of these studies would not have provided sanofi-aventis with conclusive information about the utility of clopidogrel.

Boivin J. then considered whether the promised utility of the invention was soundly predicted based on the three-part test in AZT, as modified by the Federal Court of Appeal. At the first part of the test, the Court held that “sanofi obtained results in short-term and long-term studies to support its conclusion that there was a factual basis for its prediction that the invention could be used in humans.” The “drawbacks” cited by Apotex were not sufficient to defeat the factual basis for the inventors’ prediction. At the second part of the test, the Court held that the inventors had a “reasonable” and sound line of reasoning to make a prediction of utility in humans. This was based on their knowledge of stereochemistry, toxicology, haematology, metabolism, and previous work on related compounds. Despite these findings, the Court held that the ‘777 Patent was invalid for lack of sound prediction, as it “does not instruct the POSITA that there was a factual basis and a line of reasoning for the prediction that the animal studies conducted on rat models could be extrapolated to the prediction that this compound – clopidogrel – had a use in humans.”  The Court held that the disclosure of the ‘777 Patent was insufficient because there was no reference to: (a) the work done on related compounds, or the racemate; (b) the fact that testing was conducted in multiple animals; (c) the knowledge of convulsions; or (d) the importance of metabolism to the efficacy of a pro drug (like clopidogrel). The tests disclosed in the ‘777 Patent “are with respect to only one strain of animal, in one gender (female), using only a single time point…. There was no disclosure of the basis for the POSITA to make ‘the leap’ to predict use in humans.”

Other grounds of Invalidity

The Court examined Apotex’s alternative grounds of invalidity in obiter dicta. The Court held that Apotex’s allegations of overbreadth and insufficiency were unfounded, as a person skilled in the art would have known what the claims cover and how to put the invention into practice. The Court also held that Apotex’s allegations of anticipation were not proven for two reasons: (a) the posters on which it relied were not published in the books or abstracts of any other publication and did not constitute “publications” under s. 27(1)(b) of the Patent Act (as it read prior to October 1, 1989); and (2) the published abstracts cited by Apotex did not disclose the invention claimed in the ‘777 Patent. The Court held that there was no double patenting, as the subject-matter of the claimed invention was not identical or conterminous with the invention claimed in the ‘875 Patent.

In assessing obviousness, the Court applied the four-part test that was articulated the Supreme Court in prior PM(NOC) litigation involving clopidogrel, infra. Boivin J. identified the person skilled in the art and the common general knowledge of that person. The Court held that it was “clear that there is more identified in the ‘777 Patent than what was in the common general knowledge” at the date of invention. However, Boivin J. did not consider whether these differences were “inventive”, per se (i.e., the fourth step of the obviousness analysis). Instead, the Court engaged in an “obvious to try” inquiry and asked itself the following question: “Would [it] have been self-evident to a person skilled in the art that choosing a method to separate and a method to obtain the salts ought to work.” The Court did not consider whether the “advantages” described in the ‘777 Patent would have been obvious to try. While Boivin J. held that methods to separate enantiomers and to make salts would have been obvious to try, the Court expressed no opinion on whether it would have been “very plain” that those methods would produce a compound with the advantages claimed in the ‘777 Patent. Indeed, the evidence cited by the Judge suggested the opposite conclusion.


In view of the declaration of invalidity, the Court did not need to adjudicate sanofi-aventis’ allegations of patent infringement. However, Boivin J. held that Apotex would have been liable for infringement of the product claims in the ‘777 Patent, if they had been held to be valid, based on the following acts: (1) the importation of finished goods; (2) the acceptance of purchase orders in Canada; (3) the manufacture of product in Canada; (4) the export of product from Canada; (5) the possession and use of patented goods for commercial purposes; and (6) the sale of clopidogrel or pharmaceutical compositions containing it. The Court held Apotex would also have been liable for infringement of the process claims, even if non-essential modifications had been made to the claimed process.

Boivin J. held that Apotex was not entitled to an “experimental use” defence, as it failed to provide the Court with evidence relating to what was done with the bulk material or tablets that were allegedly used for experimental purposes. In the absence of such evidence, the Court could not “rule out the possibility that all or some of the raw material or the actual formulations that were made in the course of that development process were ever used or sold for commercial purposes.” Boivin J. also held that Apotex could not rely on a defence based on limitation periods, estoppel, or abuse of process.


Apotex argued that one of the parties to the infringement action, Bristol-Myers Squibb Sanofi Pharmaceuticals Holding Partnership (the “Partnership”), did not have standing to seek a remedy, as it is not a “person claiming under the patentee” pursuant to s. 55(1) of the Patent Act. The evidence established that the Partnership was formed in 1993 to commercialize compounds on a worldwide basis and memorialized in several written agreements. The Court held that considering the “broad meaning of person claiming under the patentee” in section 55(1) of the Patent Act, and based on the Court’s review of the relevant agreements and testimony, the Partnership has a “credible and legally sufficient basis” to sue for infringement.

Prior Litigation under the PM(NOC) Regulations

In a prior proceeding under the PM(NOC) Regulations, Apotex alleged that the ‘777 Patent was invalid for anticipation, obviousness and double patenting. The Supreme Court of Canada held that these allegations of invalidity were not justified. By reason of this order, Apotex was prohibited from obtaining a notice of compliance (marketing approval) for its product until patent expiry, or until the issuance of an order invalidating the ‘777 Patent.

Sanofi-aventis relied heavily on the decision of the Supreme Court in responding to Apotex’s allegations of invalidity in the present case. However, Boivin J. held that “NOC proceedings and the fact conclusions they may have yielded are of limited assistance when, as here, the evidence adduced and the issues differ considerably from those presented in the course of the NOC proceedings.” While the Court is required to follow the legal principles and questions of law from the prior NOC case, it is not bound by factual findings that are based on a different evidentiary record. It is “incumbent upon the judge to arrive at his findings on the basis of the evidence that was before him.” Since utility and sound prediction were not at issue in the prior litigation, Boivin J. held that he was required to consider these issues de novo.

Link to decision:

Apotex Inc. v. sanofi-aventis, 2011 FC 1486