Enantiomers Are Per Se Ineligible for Data Protection

Takeda Canada Inc. v Canada (Health) / DEXILANT 2011 FC 1444 Near J  

I believe this is only the second judicial decision interpreting “innovative drug” in the Data Protection Regulations, and the first to interpret what constitutes a “variant” that is excluded from protection. In the DEXILANT decision, Near J held that the listed examples of variants in the definition of “innovative drug”, namely salts, esters, enantiomers, solvates and polymorphs, are excluded per se, regardless of whether extensive and costly trials were required to obtain an NOC.

The active ingredient in DEXILANT is dexlansoprazole, an enantiomer of lansoprazole, which had previously been marketed as PREVACID. An NOC for dexlansoprazole was issued to Takeda after a New Drug Submission “based on extensive clinical trial data generated to establish the safety and efficacy of the drug for marketing approval” [3]. Takeda requested data protection for these studies. The Minister refused on the basis that data protection is available only to an “innovative drug,” and the definition under the Regulations excludes drugs that are “a variation of a previously approved medicinal ingredient.” The Ministers’s view was that as an enantiomer of a previously approved drug, dexlansoprazole is not an “innovative drug” under the Regulations.

In arguing that dexlansoprazole should be considered an “innovative drug,” Takeda appealed to the purpose of the Data Protection Regulations. As the Regulatory Impact Analysis Statement explained, “The exclusion of variations of a previously approved medicinal ingredient from the scope of protection was introduced to avoid the granting of an additional eight years of protection where an innovator seeks approval for a minor change to a drug” (quoted at [34]). What is a “minor” change? Takeda argued that in considering whether the change was minor, the Minister should consider the nature and extent of data submitted [24]. The purpose of data protection is to provide an incentive to carry out expensive clinical trials; therefore if the clinical trials were in fact expensive, the data should be protected. Again as the RIAS explained, in determining whether a drug is a variant, “an assessment will be made as to whether or not approval is being sought primarily on the basis of previously submitted clinical data (i.e. without the support of new and significant clinical data) or not.”

The difficulty with this argument, as Near J pointed out at [32], is that the text of the definition in C.08.004.1(1) expressly excludes enantiomers:

“innovative drug” means a drug that contains a medicinal ingredient not previously approved in a drug by the Minister and that is not a variation of a previously approved medicinal ingredient such as a salt, ester, enantiomer, solvate or polymorph. (drogue innovante)

In other words, enantiomers are considered a variation per se, and accordingly there is no need to consider the nature and extent of the data submitted. The RIAS supports this interpretation. The passage quoted above refers to arguable variations not listed in the definition. The RIAS indicated clearly that all the listed variations would necessarily be considered minor and therefore excluded.

The result is that DEXILANT “therefore falls under one of the recognized variations within the meaning of subsection C.08.004.1(1), irrespective of the extent of the data collected by the Applicant as part of the NDS” [41]. Clearly the same reasoning would apply to all the listed variations, namely a “salt, ester, enantiomer, solvate or polymorph.”

It might seem that there is a conflict between the purpose and the test of the provisions. But this is not necessarily true, even if DEXILANT really did require extensive trials. This distinction between the listed variants, which are excluded per se, and other arguable variants, which are assessed in light of the nature and extent of the data submitted, brings to mind the distinction between “rule of reason” and “per se” illegality in US antitrust law. In general, the disadvantage of a per se rule is that it is sometimes over-inclusive. So, it may be, as Takeda argued, that on a full consideration of the nature and extent of the data submitted, a court would conclude that the data submitted in respect of DEXILANT should be protected in order to provide an adequate incentive to generate the “extensive clinical trial data.” But per se rules have two advantages. The lesser advantage is that it provides certainty and reduces administrative costs. The greater advantage is that is may actually be more accurate. The important point to recognize is that decision makers are not perfect, and a full review of the evidence does not guarantee a correct decision. If the great majority of listed variants are in fact minor, then it may be better to exclude them on a per se basis, and bear the downside of unfairly excluding the exceptional cases. The alternative would be to carry out a full investigation in every case, and run the risk of wrongly including variants that should have been excluded, as well as wrongly excluding some of the variants that should have been included. If the per se rule is right in the substantial majority of cases, then it can be both fairer and more efficient than a “rule of reason” approach. 

Of course, that does not mean that this particular per se rule is justified as a matter of policy. But that is a question for the legislature. It does mean that Near J's conclusion that enantiomers are excluded per se is consistent with a purposive interpretation of the provision, and his intrepretation is well justified in light of the RIAS and the text of the definition.

Finally, as a matter of policy it might also be argued that there is no need for any exclusions at all. If the variant is really so minor as to not need data protection to provide an incentive to generate the data, then it would seem that equally, the generic should not need to free ride off that data in making its submission. So, if we are to have a clear cut per se rule, then a rule that all drugs which require an NDS should also be considered "innovative drugs" is justifiable in principle. But that is clearly not the regime contemplated under the current Regulations.