Monday, January 19, 2015

A Rule Without a Principle: Patentability of Methods of Medical Treatment

AbbVie Biotechnology Ltd v Canada (Attorney General) 2014 FC 1251 Kane J
2,385,745* – anti-TNF-α antibodies / HUMIRA dosage regime

Kane J’s Humira dosage regime decision is the latest chapter in the continuing saga of the patentability of methods of medical treatment. While Kane J’s holding that the claims in question are patentable is defensible, I am not persuaded by her view that the case law is consistent [84], as there is support in the case law for the opposite result. So far as I can tell, the case law is incoherent in the sense that it cannot be explained on the basis of any unifying principle, and progress in this area of the law will not be possible until that fact is generally acknowledged.

AbbVie sells anti-TNF-α antibodies as HUMIRA for the treatment of a variety of inflammatory diseases. The application at issue in this case is for a dosage regime for the use of HUMIRA to treat those diseases [10]. The Commissioner held the claims to be novel and non-obvious. Since the use of HUMIRA to treat the various disease is in the prior art, the point of novelty lies in the dosage regime itself. The claims at issue fall into three categories: syringe claims, Swiss-type use claims and use claims [12]. (Kit claims were also rejected by the Commission, but this was not appealed [48]). Unfortunately, given that the case law seems to turn on some very fine distinctions, the actual text of the proposed claims is not provided, but it was not disputed that the essential elements of the syringe claims are [13]:

a preloaded syringe
of 40 mg of the drug Humira
for the treatment of arthritic disease or an inflammatory bowel disease;
administered subcutaneously;

Presumably the use claims were the same, except for “the use of” rather than “a preloaded syringe.” The Board held that the claims at issue are directed to methods of medical treatment and are not patentable [24]. Kane J reversed.

A preliminary question was whether the applicable standard of review of the Commissioner’s decision should be correctness or reasonableness. Kane J held that the standard of review is correctness [49]. In my view, Kane J came to the right conclusion for the right reasons, and I won’t say more about this point.

Turning to the substantive question, it is established law (at least at the FCA level) that a “method of medical treatment” is not patentable, but defining what constitutes such a method has proven to be very difficult.

Consider the following claims:

            1) use of X to treat Y (patentable)
            2) use of X in range from A to B to treat Y (unpatentable)
            3) use of X in amount A to treat Y (patentable)

It is accepted that (1) is not a method of medical treatment, and so is patentable: see Merck v Apotex / enalapril 59 CPR(3d) 133 (FCTD) aff’d on this issue 60 CPR(3d) 356 (FCA); and see CIPO’s Examples of medical use claims, Example 2.

There is case law holding (2) to be an unpatentable method of medical treatment: see Janssen / galantamine 2010 FC 1123 Barnes J; Bayer / drospirenone 2013 FC 1061 Hughes J; Axcan / ursodiol 2006 FC 527, Harrington J.

There is also case law holding (3) to be patentable: see Bayer / drospirenone 2013 FC 1061 Hughes J [159]; and similarly Merck v Pharmascience / finasteride (NOC), 2010 FC 510, [112] (though the claim at issue was to use for the preparation of a medicament). Kane J’s holding that the claims at issue in this case are patentable also evidently falls into this category ([111]-[112]), though as noted the exact claim wording is not available.

Can this pattern be explained on the basis of some principle?

A first point is that the statement that a method of medical treatment is not patentable, trite law though it may be (at the FCA level), is not particularly helpful in itself. In a very straightforward sense, the use of a compound to treat a disease is a method of medical treatment. Indeed, it is probably the paradigmatic method of medical treatment. I suspect that if you asked a random person off the street to give an example of medical treatment, he might say, “Well, I had an infection, and I went to the doctor and she gave me antibiotics, and that cured me.”

So, there is something else going on. It is now often said that the underlying concern is not to interfere with the exercise of a physician’s skill and judgment. More specifically, in Janssen / galantamine 2010 FC 1123 Barnes J stated that

[55] [T]he rationale for excluding such patents is that, for ethical and public health reasons, physicians should not be prevented or restricted from applying their best skill and judgment for fear of infringing a patent covering a pure form of medical treatment (as distinct from a vendible medical or pharmaceutical product).

While this policy seems attractive, it cannot explain the case law. The most straightforward interpretation of this rationale would be that no claim can be valid if it could be infringed by a physician in the course of their practice. But if that were the true rationale, claims in both categories (1) and (3) would also be invalid. So, if the use of AZT to treat HIV/AIDS is patented, and a patient comes to his doctor with HIV, and she prescribes AZT, she would be infringing the patent. It might be suggested that a physician would not infringe a “use” claim, on the ground that prescribing a drug is not “using” it. But if that is true, then claims in category (2) should be no more objectionable than those in categories (1) or (3).

CIPO distinguishes between a claim directed to the “what” of treatment and the “how”. The  practice notice on Examination Practice Respecting Medical Uses states as follows:

For medical inventions, the problem faced by the inventor may relate to “what” to use for treatment. Generally the solution to such a problem will be provided by an element or set of elements in a claim that embody a treatment tool. This tool may include a compound, composition, formulation, or a dosage unit form.

Alternatively, where the emphasis is not on "what" to use but instead relates to "how" to administer or refine a treatment, the solution (as embodied by the essential elements of the claim) will likely place a limit on the professional skill or judgment of a physician. An emphasis on "how" (as distinct from "what") may include details of when or where a treatment is to be administered or who is to receive a treatment.

The accompanying Examples distinguish between category (1) as being statutory and category (2) as being unpatentable on the basis that the former is directed to the “what” and the latter to the “how” (compare Examples 2 and 4). I appreciate that CIPO is trying to make sense of a difficult body of case law, and this provides at least a verbal distinction between the categories. But if a patient comes in with HIV, and the physician’s view is that the best treatment is 2.0 mg of AZT, the physician’s skill is just as limited whether there is a patent on the use of AZT to treat HIV, or on the use of AZT in the range of 1 - 5 mg to treat HIV, or on the use 2mg of AZT to treat HIV. Whichever way you slice it, a patent over a range doesn’t restrict the physician any more or less than a patent over the use of the drug, or over a specific dosage.

Certainly, if a physician would have prescribed the drug in question (in the context of a category (1) patent), or a dosage in specified range or amount (category (2) or (3)), even prior to the disclosure of the claimed invention, then the invention would be an undue restriction on the exercise of their skill and judgment. The AG argued as much in this case:

[82] The respondent also suggests that physicians may have been prescribing Humira in a bi-weekly dosage and by patenting the dosage claimed, the exercise of the physicians’ skill will be restricted and the existing prescribing practice will risk infringing the patent.

But that cannot be true, or the patent would be anticipated or at least obvious, which is contrary to the Patent Office’s express determination with respect to the patent at issue in this case [4]. More generally, if the patent is otherwise valid, we must take it as established that a physician, in the ordinary exercise of their professional skill, would never have practiced the claimed invention prior to it being disclosed by the patentee.

It is also said that the claims are invalid if they “rely upon” the skill and judgment of a physician in some sense ( 2010 FC 510 [114]). So, in the PAB decision in Allergan, CD 1292, (2009) 79 CPR (4th) 161, apparently approved by Kane J as correctly stating the law [88], the Board stated:

if a dosage is claimed as part of the patent monopoly it must not be in the form of a range, such that in order to determine the appropriate dosage for a particular patient, specific knowledge of that patient is required, and judgment is required based on that knowledge, matters which fall within the skills of the physician, and are therefore unpatentable.

At first glance this does seem to distinguish between categories (2) and (3): a specific dosage regime, because it is so specific, requires no skill and judgment. This is the point evidently accepted by Kane J in distinguishing this case from precedent in category (2).

There are two problems with this explanation of the case law. First, I entirely agree with Kane J’s point that the administration of any drug requires the exercise of skill and judgment:

[110] The respondent’s approach [that a fixed dosage claim is unpatentable] appears to overlook that the professional – in this case the physician – must still exercise their skill and judgment to decide whether or not the claimed invention should be used as claimed (i.e., whether it is appropriate for a particular patient).

But if that is the true, then why are claims in category (2) objectionable? They too require a judgement as to whether the claimed invention is appropriate for a particular patient.

To answer this question, Kane J seems to have relied on a distinction between skill and judgment in deciding whether the drug should be administered, and skill and judgment in selecting the dosage:

[110] Once that is determined, there may be no further need to exercise skill and judgment to vary the claimed way to use the invention. It is not expected and it is not necessary. The expert evidence about the invention will inform whether skill and judgment is expected or necessary to be exercised.

[111] In the present case, there is no evidence to suggest that the bi-weekly dosage is not fixed and precise and that further skill and judgment would be expected to be exercised.

I find this distinction very difficult to accept. If the administration of any drug requires the exercise of skill and judgment, what does it matter whether that judgment must be exercised in the determination of whether the drug is required at all, or whether it is exercised in the determination of the exact dosage? Yes, a distinction can be drawn between skill in determining whether a particular patient needs a drug, and skill in determining how much they need, but what is the principle underlying this distinction? Essentially this point which was made by Hughes J in Zoledronate 2013 FC 985 aff’d 2014 FCA 17 in holding all the claims in question to be invalid:

[94] The '201 patent specifically states that the mode of administration and dosage “may be selected by the attending physician taking into account the particulars of the patient, especially age, weight, life style, activity level, hormonal status (e.g. postmenopausal) and bone mineral density as appropriate”. (page 11, Hughes J’s emphasis)

[99] [B]ecause each claim of the '201 patent, directly or by incorporation by reference, includes as well treatment by intermittent dosages with some claims specifying a dosage range and others specifying specific dosages; and some claims claiming more frequent intervals of dosing, and others less; that the claims include that which lies within the skill of the medical practitioner and are thus invalid.

The claims held invalid by Hughes J included Claim 16 of the ‘201 patent, which construed with the most restrictive dependencies, is as follows:

16. Use of [zoledronate] for treatment of osteoporosis, administered annually, intravenously, in a dosage of 5 mg.

This restrictive construction was explicitly recognized by Hughes J at [70], and he nonetheless held the claim invalid. That holding is very difficult to reconcile with Kane J’s holding in this case. That is not to fault Kane J, as it is also very difficult to reconcile with Hughes J’s own holding in Drospirenone 2013 FC 1061, in which the following claim was held valid and not directed at a method of medical treatment [159]:

8. Use according to claim 1, whereby the estrogen is present in a dose of 20 µg of ethinylestradiol or an equivalent dose of 17-estradiol and the gestagen is present in a dose of 75 µg of gestodene or an equivalent dose of levonorgestrel, cyproterone acetate or drospirenone.

In Drospirenone Hughes J held all the claims invalid, except for claim 8, because “they provide for a choice to be made by those prescribing or providing contraceptive drugs to choose between a variety of components and a variety of dosage ranges. Only claim 8 survives, as it is directed to a single dosage of each of two compounds” [162]. Kane J’s reasoning in this case is consistent with Hughes J’s reasoning in Drospirenone, but it is inconsistent with his reasoning in Zoledronate.

A second problem with the notion that claims are unpatentable if they rely on the exercise of a physician’s skill, is that this view is impossible to reconcile with the patentability of of category (1) claims, which will invariably rely on the skill and judgment of the physician to determine the appropriate dosage for a particular patient. The claim at issue in Wellcome / AZT 2002 SCC 77, for example, was to a formulation comprising “an effective amount” of AZT ('277 claim 22). In the context of a use claim without any specific dosage regime, that effective amount must be determined by the physician in the exercise of their skill and judgment. It is no doubt ordinary skill and judgment of a physician, or the patent would be invalid for insufficiency, but it is medical skill and judgment nonetheless – I have no idea whether the 2.0mg of AZT I suggested in my example above would be an effective treatment for HIV, but no doubt a competent physician could tell me. A claim to the use of X to treat Y cannot possibly involve less skill and judgment than a claim to the use of X in range from A to B to treat Y.

It might be said that what is different is that in the case of category (2), an open-ended dosage regime, is that the inventive concept of the claim is something that would normally be within the exercise of the physician’s professional skill. But this does not successfully distinguish category (1). For example, there are myriad antibiotics currently on the market. It is a quintessential example of the exercise of a physician’s skill to decide which particular antibiotic is useful in treating a particular disease for a particular patient. The claim to the use of compound X to treat disease Y is exactly something that would be an exercise of the physician’s skill. Moreover, I intentionally slipped in the word “normally” to make the point sound more plausible. We must keep in mind that if the claimed invention in any category is valid, the particular treatment in question would not normally have been prescribed by the physician. It is much less compelling to say that a physician should not be restrained in the exercise of a skill which they would never have possessed without the disclosure of the patent.

At the end of the day, I simply don’t see any principled way to reconcile the various decisions. Of the various suggestions in the case law, the only one that strikes me as being truly principled is that physicians should not have to fear patent infringement lawsuits when treating their patients. But, as discussed, that cannot explain the cases, given that second medical use claims threaten doctors as much as dosage regime claims. Moreover, it would wrong be to rationalize the cases by holding by holding second medical use claims unpatentable. That would be bad policy in my view, given the importance of such inventions, as is illustrated by the use of AZT for the treatment of HIV/AIDS; but more importantly, it would be such a momentous change that it must be left to the legislature. Moreover, there is no particular reason to believe that such a rule is necessary. Certainly it is reasonable to say that physicians should not fear patent infringement actions when treating their patients, but so far as I understand, physicians in Canada are not the target of lawsuits by patentees, and as a practical matter it seems unlikely that brand pharmaceutical companies would want to antagonize their target market. The real threat in that regard is more likely to come from patents related to surgical techniques, which are patented by individual physicians, and for which there is no upstream infringer to target apart from the infringing surgeon herself: that was the context which led to the legislative defence in the US (see eg 35 USC § 287(c)(1)). It is overbroad to respond to the possible threat posed by surgical technique patents by targeting patents for pharmaceutical dosage regimes, where the litigation is invariably against a generic pharmaceutical supplier. If such a defence is required, it should be done directly, by creating a defence for physicians, whether by statute or through the case law, rather than by torturing the bounds of patentable subject matter.

The root of the problem is the SCC’s decision in Tennessee Eastman, which suggested that methods of medical treatment are unpatentable in a single throwaway line of a decision that was substantively concerned entirely with s 41. The difficulty was compounded by the SCC in Wellcome / AZT 2002 SCC 77, [49]-[50] when it “explained” Tennessee Eastman in two short paragraphs that managed to give five or six vague and contradictory rationales, in addition to the straightforward and sufficient holding that Tennessee Eastman turned on s 41. Unfortunately, the view that Tennessee Eastman established that methods of medical treatment are unpatentable was accepted by the FCA in ICI [1986] 3 FC 40, 9 CPR(3d) 289 (FCA), but solely as a matter of the authority of Tennessee Eastman, without any principled explanation of that prohibition in the absence of s 41.

Without any principled basis for the rule, it is no wonder its application is confused. The confusion is fundamental. As soon as we state the rule that in Canada, “methods of medical treatment” are unpatentable, we have to go on explain that “the use of compound X to treat disease Y” is not really a method of medical treatment. So far as I can tell, the real reason for the current focus on certain types of dosage regimes is that the courts are desperate to give some content to the rule that methods of medical treatment are unpatentable, and dosage regimes in some vague way seem to fit the bill, without being so fundamental as to be beyond attack. Yet even in the narrow context of dosage regime, there is no clear rule, which is a direct consequence of the fact that there is no sound principle underpinning the rule. It should be no surprise that an edifice built on quicksand is so shaky.

*The decision states that the application at issue is 2,835,745 [1], but this evidently is a typographical error, with the 3 and 8 reversed.


  1. I’ve wondered whether some sense might be made of some of the cases by distinguishing between claims that require professional skill and judgement to arrive at an inventive dosage regime, as opposed to claims that cover subject matter that necessarily encompasses the exercise of skill and judgement but where that skill and judgement will always be exercised within the scope of an overarching inventive contribution. For example, a “use of X for treating Y” claim necessarily covers the use of professional skill in selecting a particular dosage regimen, but you could argue that no inventive contribution is required to adapt the dosage regimen to the newly discovered treatment. In contrast, claims that cover a titration regimen leave a great deal in the hands of a proscribing physician, who must exercise judgment in an attempt to achieve an effective titration. This may be a new use, i.e. effective titration to a lower final dosage, but the route to achieve that new use is one that arguably requires the exercise of inventive professional skills for success. One can imagine a basis in s.2 for recognizing an invention only when inventive professional skills are not required to make and use the invention. This ‘enablement’ line of argument is of course not the direction the cases have taken, and even this nuanced approach doesn’t fit all the cases.

  2. Hi Anon,
    I think I see where you are going, but you have pointed out the main weaknesses with this line of thought, which is that it is an enablement argument, and does not accord with what the courts say they are doing, nor does it explain the actual outcomes on the facts. So, you are saying that the courts may be applying an enablement analysis in the guise of a s 2 analysis, and doing a poor job of it. That is possible – and if they are actually carrying out an enablement analysis, the fact that they are hearing argument on s 2 would explain why they are doing it so poorly. But if that were the case, I would expect to see at least a few hints that enablement-type considerations had swayed the outcomes, and I just don’t see that in the cases. I think the simpler explanation is that they are trying to do a s 2 analysis, and doing a poor job of that because there is no underlying principle.

  3. Hi Norman,
    I wasn't trying to do the perhaps impossible job of making sense of the decisions; I was suggesting an alternative form of analysis that might get you to somewhere like the same place. But, as you say, that's not the path the courts have taken. What we have is a morass of logical inconsistencies, as you so politely highlight.

  4. Hello. Different Anonymous here (eek, did someone say "contribution"?).

    This is one of the most confusing and nonsensical aspects of Canadian patent law. I used to be able to wrap my head around when things were simpler, i.e., when "method" claims were rejected and "use" claims were generally accepted. Now that narrow "uses” (e.g. dosage ranges) are falling and deemed to be objectionable by the Patent Office (e.g. uses in patient subpopulations), it makes no sense whatsoever.

    It is also interesting to note that the language of Patent Office objections has shifted away from the "essentially non-economic" language of Tennessee Eastman that used to be quoted, and towards the "professional skill" language drawn ultimately, I think, from Lawson.

    I did some reading about this recently, and it appeared to me that Lawson was cited to justify a validity attack in the 2001 Visx v. Nidek decision. In obiter, the court concluded that the use of the laser eye surgery machine did not inhibit physicians (I’m paraphrasing). Thereafter, this obiter was cited to justify an anxiety about inhibiting the discretion physicians, which has propagated subsequently and resulted in a conflation of issues.

    Lawson says only that "...professional skills are not the subject matter of a patent. If a surgeon were to devise a method of performing a certain type of operation he cannot obtain and exclusive property or privilege therein.” This is a long way from saying that a patent claim cannot ever inhibit a physician.

    On the subject of professional discretion itself, the term “discretion” implies that there is judgement to be exercised that a patent could inhibit. I would argue that there is no discretion to exercise if the information about a new dosage range or patient subpopulation is not known. And it’s not as if physicians are in the habit of fine-tuning all medications, as they are, say, a morphine drip. Often, only set dosage forms are approved or available, and many of these are not amenable to subdivision due to, e.g., coatings and controlled release properties. Discretion can be limited to a choice between 20mg and 40mg. Likewise, most family physicians are not int habit of carrying out detailed pharmacogenomic population studies to determine which patient sub-groups might react more positively or negatively to a certain drug that is already prescribed generally. In short, the anxiety doesn't even reflect the normal activities of physicians!

    These discoveries are not within the purview of professional skill or judgement, but that of of medical research. They should be fair game for patent protection as far as I'm concerned, and the effects of disincentivizing this industry should not be taken lightly.

    I feel a CIPR article coming on...