Federal Court opines on precedential value of prior PM(NOC) proceedings and scope of experimental use exception to public disclosure

Ryan Sheahan and Jessica Rutledge - 

The Federal Court’s decision in the consolidated proceedings of Bayer Inc. v Cobalt Pharmaceuticals Company  offers useful guidance as to the precedential value of proceedings under the Patented Medicines (Notice of Compliance) Regulations (PM(NOC)) in the context of subsequent infringement actions involving the same patent.  It also provides helpful direction in respect of the scope of the Canadian experimental use exception to anticipatory public disclosure.  

Background:  Patent at Issue

Both the PM(NOC) proceedings and the Federal Court infringement action concerned Bayer’s Canadian Patent No. 2,383,426 (the ‘426 Patent).  The ‘426 Patent relates to a pharmaceutical combination of ethinylestradiol and drospirenone for their use as a contraceptive.  Ethinylestradiol functions as an estrogen, and drospirenone as a progestogen, and both were known to inhibit ovulation when used in combination.  The ‘426 Patent contemplated a novel formulation using rapidly-dissolving drospirenone particles, which delivered unexpectedly consistent bioavailability.  This was unexpected because drospirenone was known to be inactive at an acidic pH of 1 when tested in a laboratory setting, and the normal pH range of the stomach is 1.0 to 3.0.  However, traditional coatings that would protect the drospirenone from the stomach environment released the drospirenone at variable rates, resulting in irregular effectiveness in different people.  Bayer tested a rapidly-dissolving particulate version of drospirenone in clinical trials with women, and discovered it was effective as a contraceptive in the ethinylestradiol combination, and did not degrade or require any coating.  Bayer patented this formulation of the contraceptive in the ‘426 Patent, with an effective Canadian filing date of August 31, 2000, and currently markets two products known as “Yaz” and “Yasmin”, which incorporate the patented invention, with different doses of ethinylestradiol .

Background:  Previous PM(NOC) Proceedings and New Infringement Action

In Canada, a generic drug manufacturer can seek approval to market (a “Notice of Compliance” or NOC) a generic version of a brand name drug by filing an Abbreviated New Drug Submission (ANDS) with Health Canada.   An ANDS cross-references the submissions for the approved brand name drug and includes bioequivalence studies and data to establish that the generic product is pharmaceutically equivalent.  Such bioequivalence studies generally cost less and take less time than the clinical trials required to be conducted to obtain the initial approval for the brand name drug.

The PM(NOC) allow brand name manufacturers to list one or more patents it may have that claim any aspect of the brand name drug (e.g. the medicinal ingredients, the dosage form, formulation, and/or uses) on Health Canada’s Patent Register.  Pursuant to the PM(NOC) Health Canada cannot approve a generic drug product until the expiry of the last listed patent, unless the generic manufacturer has served a notice of allegations on the brand manufacturer challenging the validity of one or more listed patents and/or alleging that the generic product does not infringe such patents.  The brand manufacturer has the option to bring an application to the Federal Court seeking an order to prohibit Health Canada from approving the generic drug product on the basis that the allegations of invalidity or non-infringement are not justified.  If the Court concludes that the allegations of invalidity or non-infringement are justified, Health Canada can approve the generic drug product.

Bayer listed its ‘426 Patent on the Patent Register against its Yaz tablets.  Both Apotex and Cobalt sought regulatory approval of their generic versions by comparison to the Yaz tablets and serving notices of allegations on Bayer with respect to the ‘426 Patent, in two PM(NOC) proceedings:

  • In December 2011, Cobalt sought approval for its proposed generic version of the Yaz tablet.  In seeking a Notice of Compliance, Cobalt alleged both that its drug did not infringe Bayer’s ‘426 Patent, and that the ‘426 Patent was invalid.  Bayer was successful in its application to prohibit the Minster of Health from issuing a Notice of Compliance in Bayer Inc v Cobalt Pharmaceuticals Company (Bayer v Cobalt), which was then upheld by the Federal Court of Appeal. [Cobalt Pharmaceuticals Company v Bayer Inc (Cobalt v Bayer)].
  • In July 2012, Apotex initiated similar proceedings for approval of its generic version of Yaz, claiming both non-infringement and invalidity of the ‘426 Patent.  On Bayer’s application to prevent Apotex’s Notice of Compliance, the Federal Court again upheld the validity of the ‘426 Patent, but found Bayer had not met its burden of proving that Apotex’s product fell within the scope of the ‘426 Patent’s claims (i.e., the allegations of non-infringement were found to be justified).  [Bayer Inc v Apotex Inc ] (Bayer v Apotex) Apotex subsequently obtained approval of its generic Yaz, and was able to bring the product to market.

Following these PM(NOC) proceedings, Bayer commenced three separate actions for infringement, one against Cobalt for the manufacture and sale of its Yasmin generic, and the other two against Apotex for the manufacture and sale of its Yaz and Yasmin generics, all of which were consolidated by the Federal Court.  Both Cobalt and Apotex denied the infringement allegations and, by counterclaim, alleged the ‘426 Patent’s claims were invalid for obviousness, anticipation, disclosure, overbreadth, insufficiency and lack of utility.  Bayer was successful in upholding the validity of the ‘426 Patent, and the Federal Court found infringement in all three cases. 

Federal Court Decision:  Precedential Value of the Prior PM(NOC) Proceedings

In coming to its decision in respect of the infringement claims and counterclaims for invalidity, the Federal Court provided a detailed analysis of the precedential value of prior PM(NOC) proceedings involving the same patent.  At the outset, the Court confirmed that the principles of stare decisis and comity apply where courts are exercising the same jurisdiction, for instance in PM(NOC) proceedings involving a patent previously considered in the PM(NOC) context.  However, courts considering a subsequent action for infringement are not exercising the same jurisdiction as courts that made findings in the PM(NOC) context.

The Federal Court noted that PM(NOC) proceedings are summary in nature, address different issues than trials, and use only affidavit evidence.  As such, they cannot result in a final determination of validity or infringement.  However, certain findings will be either prima facie binding, or potentially persuasive:

  • Questions of law:  pronouncements of higher courts on questions of law are binding on lower courts.  In the patent context, the construction of claims is treated as a question of law, and therefore the principle of stare decisis applies.  In this case, the Court found that the Federal Court of Appeal’s prior construction of the ‘426 Patent in Cobalt v Bayer was prima facie binding, but did note that the construction could be revisited if warranted by new evidence.  Claims construction includes the definition of the patent’s “inventive concept” and its “promise”.
  • Questions of fact or mixed fact and law: previous findings involving fact-based inquiries are potentially persuasive, but should be determined anew based on the evidence adduced at trial.  The Court confirmed that the determination of the person of ordinary skill in the art, as well as assessments of obviousness, ambiguity, overbreadth, utility and insufficiency were all questions of mixed fact and law. 

In this particular case, the Federal Court did not choose to depart from the Federal Court of Appeal’s construction of the claims.  Nor did it depart from any of the Federal Court of Appeal’s findings regarding the validity of the ‘426 Patent, in respect of the three grounds for invalidity (of the six argued at trial by the defendants) that were previously argued in the PM(NOC) proceedings.

As noted above, the Federal Court found infringement in all three of the joined cases.  In doing so, the Court agreed with the Federal Court of Appeal’s infringement analysis in the Cobalt v Bayer case, but the result differed from the Federal Court’s decision in the Bayer v Apotex case.  Essentially, the infringement claims turned on which expert evidence the Court accepted, in the context of a far more extensive inquiry at trial than was undertaken in the PM(NOC) context.

Federal Court Decision:  Regulatory Approval Trials Constitute Experimental Use

The case also provides useful analysis of the experimental use exception to anticipatory public disclosure.  Pursuant to s. 28.2 of the Patent Act, if a patentee discloses the subject matter of a patent to the public over a year before the patent’s filing date, to a degree that would enable the skilled person to practice the invention, the patent is invalid for anticipation. 

More than one year before the filing date, Bayer conducted clinical studies of oral contraceptives as indicated by the ‘426 Patent.  Apotex and Cobalt argued that these clinical studies constituted an enabling disclosure.  The Federal Court did not find these trials constituted an enabling disclosure, holding that reverse-engineering the tablets would have required inventive insight to discover the inventive concept.  In addition, the Court noted that even if the trials did constitute disclosure, the experimental use exception would apply.  Though not codified in the Patent Act, Canadian law has long held that there is no anticipatory disclosure where a prior use is experimental, and that an inventor may use any means of testing available to him or her that is reasonable and necessary, and done in good faith for the purposes of perfecting or testing the merits of an invention.

In this case, the Federal Court agreed with the application of the experimental use exception as applied in Bayer v Apotex,which noted the exception’s particular applicability in cases where public testing is a necessity.  In the case of oral contraceptives, clinical studies are necessary to prove a drug is safe and effective, and to gain government approval for sale, and both are compulsory steps for commercialization of the drug.  Apotex argued that the last round of clinical trials could not be experimental as they were undertaken for governmental approval and not to prove the drug was safe and effective.  However, the Federal Court found that such a purpose did not bring the trials outside the experimental use exception, as one of the purposes of regulatory approval is to confirm the safety and efficacy of a drug.

Next Steps

The issue of appropriate relief was bifurcated from the issues of validity and infringement.  On October 27, 2016 in Bayer Inc v Cobalt Pharmaceuticals Company,the Federal Court ordered that Bayer could was entitled to elect between an accounting of the profits of Apotex and Cobalt, or all the damages sustained by their infringement of the ‘426 Patent.  

Both Cobalt and Apotex have appealed the Federal Court’s decision.  

Court of Appeal clarifies - for utility the promise of the patent must be "clear and unambiguous"

Ryan Sheahan -

On July 24, 2013, the Federal Court of Appeal (FCA) issued its decision in Apotex Inc. v. Sanofi-Aventis, overturning a Federal Court decision invalidating Canadian Patent No. 1,336,777 (’777 Patent) - a patent owned by Sanofi claiming the active ingredient of the drug Plavix (clopidogrel bisulfate). This decision is the latest step in a lengthy history of proceedings between the parties, including an appeal to the Supreme Court of Canada (SCC), which held that the ’777 patent was valid. As that SCC decision was in the context of the Patented Medicines Notice of Compliance (PM(NOC)) proceeding, the finding was not (strictly speaking) binding on the Federal Court. It was still surprising that the Federal Court subsequently decided to invalidate the patent. The FCA decision points out the errors of the Federal Court and clarifies the law on the promise of the patent.

In order for an invention to be patentable in Canada it must be new, inventive and have utility. (This is true under both the pre-October 1, 1989 Patent Act (the Old Act, which the ’777 Patent was subject to) and the post-October 1, 1989 Patent Act.) Generally speaking, the threshold for meeting the utility requirement is quite low; absent a promise in the patent, a “mere scintilla” of utility will suffice. On the other hand, when a patent sets out an “explicit” promise, the utility of the invention will be assessed against such a promise. If the invention does not meet the stated promise – the patent will fall for lack of utility.

The FCA reiterated that not every patent contains an explicit promise of utility and that there is no legal obligation to disclose the utility of the invention in the patent. Selection patents are an exception to this disclosure requirement. The utility requirement of a selection patent is met by the selected compounds having an unexpected advantage over the other compounds in the genus and by disclosing such unexpected advantages in the patent. The ’777 Patent is considered to be a selection patent because it claims a subset of compounds that fall within the scope of the teachings and claims of an earlier genus patent. 

The trial judge concluded that the ’777 Patent contained an “explicit promise” regarding the use of the compounds in humans (i.e., that the compounds will be useful in humans). The trial judge came to this conclusion by drawing inferences from various statements in the patent, for example, the references to the “medicine of the invention” and that the “medicine of the invention can be usefully administered in the treatment and prevention of platelet disorders…”.

The FCA held that the trial judge erred in law by reading this promise into the patent. The FCA stressed that courts should not strive to find ways to defeat otherwise valid patents. In other words, the courts should avoid reading a promise into a patent where one does not explicitly exist.

In reaching its decision, the FCA emphasized the distinction between the potential use of an invention and an explicit promise of a specific result, noting that inferences, goals, hopes and potential applications are not promises. The FCA concluded that the trial judge “erred in law in reading into the ’777 Patent a promise for use in humans on the basis of inferences, in the absence of language at least as clear and unambiguous as that used to establish the advantages of the selection over the compounds of the genus patent” (emphasis added). The “clear and unambiguous” statements referred to by the FCA include statements such as “[i]n an unexpected manner only the [selected] enantiomer exhibits a platelet aggregation inhibiting activity, the [unselected] enantiomer being inactive. However, the inactive [unselected] enantiomer is the less well tolerated of the two enantiomers.”

In other words, the FCA concluded that the patent did not promise that the compounds would be useful in humans. According to the FCA, all the patent promised was that the selected compounds had greater platelet aggregation inhibiting activity and were less toxic than the unselected compounds. The patent had data to support such a promise. Therefore, the patent was not invalid for lack of utility.

Justice Gauthier, in his concurring decision, also noted that Canadian applications filed on the basis of European applications will often contain statements regarding the invention’s industrial application because European patent law requires the invention be capable of industrial application. Justice Gauthier cautions that “one must be careful not to treat each reference to a practical purpose as a promise of a specific result…”.

Though it is still possible for Apotex to seek leave from the SCC to appeal this decision, it is unlikely that such leave will be granted. The SCC has already issued a decision regarding the validity of this particular patent, albeit in the context of a PM(NOC) proceeding, and this particular utility issue was not raised. The SCC may be reluctant to have another hearing to again decide the validity of this patent.

On a strict reading of the FCA’s decision, absent an explicit promise, i.e., a clear and unambiguous statement, it will be very difficult to find that a patent promises any particular utility. Absent an “explicit promise”, all the patentee must show is that the invention has the merest scintilla of utility.

Federal Court of Appeal decision clarifies effects of partial waiver of privilege

On January 30, 2009, the Federal Court of Appeal (FCA) clarified the circumstances under which a partial waiver of privileged material can result in an implied complete waiver of privilege. The FCA's decision in Merck & Co. Inc. et al. v. Apotex Inc. et al. (2009 FCA 27) was an interlocutory decision in a patent infringement case.

The issue arose in the context of examination for discovery in respect of the patent dispute, which relates to the manufacture of a drug called lovastatin. This drug is produced by certain micro-organisms, and the production of lovastatin from the Aspergillus terreus micro-organism is covered by a Canadian patent owned by the plaintiff, Merck & Co. Inc. (Merck). Merck had apparently carried out some tests, for the purpose of the litigation, to improve its understanding of the process used by the defendants. As the tests were conducted for the purpose of the litigation, information about the tests qualified for protection under the litigation privilege. The purpose of the litigation privilege is to create a zone of privacy in relation to pending or apprehended litigation, to facilitate investigation and preparation of a case for trial by the adversarial advocate.

At an examination for discovery, Merck's representative answered certain questions relating to these tests. Apotex then asked further questions regarding the tests, which the Merck representative refused to answer, asserting that the information was privileged.

It is clear that before the Merck representative had answered any questions about the tests, the entirety of the information was privileged, and Merck did not have to answer questions about them. However, when Merck answered certain questions about the tests, this privilege was partially waived. When a partial waiver of privilege has occurred, the law requires that the privilege in issue be completely waived if permitting some of the information to remain privileged would lead to inconsistency or unfairness.
This rule exists because the purpose of privilege is to preserve a zone of privacy for certain types of information. Privilege does not exist to enable litigants to unfairly and inconsistently disclose and rely on the favourable elements of these types of information, while maintaining the secrecy of the unfavourable elements. 

In overturning the Federal Court decision, the FCA took issue with the lack of identification of any particular inconsistency or unfairness that might arise if the partial waiver of privilege stood. The FCA noted that the Federal Court decision simply invoked the concepts of consistency and fairness, but did not explain them.

In contrast, the FCA carefully examined possible unfairness to the defendants, and found that no unfairness could result. Merck could not make use if its own partial responses, and the test information itself, if it were to be relied on, would have to be disclosed well in advance of trial. Having found no actual unfairness to the opposing party, the Court of Appeal held that fairness and consistency did not require the complete waiver of privilege. Accordingly, the additional information about the test, sought by Apotex, could remain undisclosed.

The case is a reminder of the care that must be taken when issues of privilege arise. Partially waiving privilege is a significant strategic decision, and the possible ramifications of such a decision must be carefully considered.