Is a Patent Insufficient If it Requires a Minor Research Project to Practice?

Takeda Canada Inc v Apotex Inc 2024 FC 106 Furlanetto J

2,570,916 / dexlansoprazole / DEXILANT / NOC

As discussed in my last post, the patent at issue in this case related to a “pulsatile” dosage form of proton pump inhibitors (PPIs) comprising a PPI with “a first and a second dose,” which are released from the dosage form as “discrete pulses,” resulting in specified blood plasma concentrations [90], [92]. Takeda’s DEXILANT product is a pulsatile release formulation of the claimed type, which includes two types of delayed-release beads containing dexlansoprazole [6]. Apotex sought to sell a dexlansoprazole oral dose capsule product and Takeda brought this NOC action in response [2]. As discussed in my last post, Furlanetto J held that Takeda had not established infringement. That aspect of her decision was entirely straightforward, and would have been sufficient to dismiss the action [148]. Furlanetto J nonetheless went on to address Apotex’s validity arguments, “which formed a significant portion of the parties’ arguments at trial” [148]. The last post discussed the novelty argument. This post deals with the remaining validity issues. It raises the thorny issue of whether the factual basis for a sound prediction must be disclosed in the patent; I review the debate, though in the end this decision adds nothing new, as the point was not contested. The other interesting issue is the treatment of sufficiency, which strikes me as problematic in apparently requiring a patentee to disclose in the patent the amount of active ingredient necessary for clinical efficacy.

Obviousness

There is nothing notable in the obviousness analysis, but a brief description is useful in understanding the utility argument. While pulsatile dosage forms were known in the prior art for dealing with the breakthrough effect, there were some differences between the prior art and the asserted claims: in particular, the asserted claims specified the plasma concentrations needed to prevent breakthrough [209] and also required the second dose to be larger, to compensate for reduced absorption in the lower digestive tract [213]–[215]. Furlanetto J held that these differences would not have been obvious, in an analysis that turned on the facts [217]–[227].

Utility

Furlanetto J stated that “where the utility is founded on a sound prediction, the factual basis for the prediction must be set out in the patent disclosure” to the extent it is not based on the CGK [231]. Whether the factual basis for a sound prediction needs to be disclosed in the patent is a long running debate seeming from an enigmatic statement in Wellcome / AZT 2002 SCC 77 [70]:

Thirdly, there must be proper disclosure. Normally, it is sufficient if the specification provides a full, clear and exact description of the nature of the invention and the manner in which it can be practised. . . . In this sort of case, however, the sound prediction is to some extent the quid pro quo the applicant offers in exchange for the patent monopoly. Precise disclosure requirements in this regard do not arise for decision in this case because both the underlying facts (the test data) and the line of reasoning (the chain terminator effect) were in fact disclosed, and disclosure in this respect did not become an issue between the parties. I therefore say no more about it.

The Federal Courts initially understood this as requiring only the standard disclosure of how to make and use the invention: see eg Aventis Pharma v Apotex 2006 FCA 64 [28]–[35] affg 2005 FC 1283 [178]–[254]. Then, in Raloxifene 2008 FC 142 [164] affd 2009 FCA 97 [15], Hughes J interpreted this paragraph of Wellcome / AZT as requiring disclosure of the factual basis for the sound prediction in the patent itself. (There’s also a separate issue as to what is meant by “this sort of case”: see AstraZeneca FC 2014 FC 638 [141], discussed here.)

There are two main problems with this. First, the factual basis for sound prediction was not disclosed in the patent in the leading cases: see my post “The underlying facts were NOT in fact disclosed,” showing that the factual basis for the sound prediction was not disclosed in Wellcome / AZT itself; and see my comments to that post showing that the factual basis was not disclosed in Olin Mathieson [1970] RPC 157 (Ch) either, which Wellcome / AZT [60] identified as the case which gave “serious shape and substance” to the doctrine. Second, it seems clear that there is no duty to disclose the factual basis for demonstrated utility in the patent. (Initially it was entirely clear that there was no such duty, but after the enhanced disclosure requirement was advanced for sound prediction, there have been some hints that there is an enhanced disclosure requirements for demonstrated utility as well.) There is only one utility requirement in the Act, and it is often a very fine line between demonstrated utility and sound prediction — the difference will turn on exactly how much data has been collected. So, if the facts A, B, C and D are enough to demonstrate utility, but the facts A, B & C are only enough to establish a sound prediction, if the applicant is in possession of the facts A, B, C and D, it will not need to disclose any of them, but if the applicant is only in possession of the facts the facts A, B & C, they will need to be disclosed. It is difficult to see any rationale for this distinction. The only rationale that has been provided for the heightened disclosure requirement for sound prediction is the statement in Wellcome / AZT [70] that “the sound prediction is to some extent the quid pro quo the applicant offers in exchange for the patent monopoly.” This statement is completely opaque, which is entirely understandable, given that it was expressly obiter. As a result, it is not clear what the rationale is for a heightened disclosure requirement for sound prediction; and it is double unclear as to what the rationale might be for distinguishing the disclosure requirement in a case of sound prediction from that when utility is demonstrated.

In this case, the parties evidently accepted that the factual basis must be disclosed in the patent, as Furlanetto J did not indicate that her statement at [231] was disputed. I’m not convinced that the the point is settled. Furlanetto J relied on Eurocopter 2013 FCA 219 [153], Apotex v Allergan 2015 FCA 137 [9] and Pharmascience v Teva 2022 FCA 2 [5]. Eurocoper says only that disclousre of the factual basis “may” be required and Apotex v Allergan merely paraphrased this in a brief decision affirming the decision below on the facts. Pharmascience v Teva 2022 FCA 2 is more directly on point, but even there Locke JA’s remarks are ambiguous, as discussed here. Since the parties did not dispute the point, this decision doesn’t add any authority to the debate. The issue will eventually have to be clarified by the FCA.

Given that position, the question was whether two examples provided a sufficient factual basis for a sound prediction of utility. The parties agreed that the requisite utility was simply “to give a pharmacological effect (i.e., as agreed by the parties in this context, an effect on gastric acid pH)” [233]. Furlanetto J found that a scintilla of utility had not been established on the facts [248], apparently because “the effect of a pulsatile release dosage form on gastric pH could not have been predicted” [244].

I have to admit, I find this very suprising. It is very well known that PPIs are effective in treating GI issues, at least when given in a large enough dose to meet threshold blood plasma levels. Claim 11, which was one of the assserted claims [2], specifies blood plasma levels of at least 450 ng/ml, which I take it is well over the threshold. Presumably if this concentration were achieved with a single dose, it would be effective. I find it very difficult to understand how it makes any difference if the same plasma level is achieved with pulsatile dose. But I have to admit that I couldn’t really follow the technical evidence reviewed by Furlanetto J, and anyway, it is a finding on the evidence.

Sufficiency

Furlanetto J held the disclosure was insufficient for two reasons.

[258] First, Example 1 does not provide sufficient information for the PSA to understand how the inventors arrived at the oral dosage data and steady state plasma concentrations in the patent. As highlighted earlier and explained by Dr. Davies, the experimentation said to underlie Example 1 yielded modelled results that would be known to the PSA to be significantly lower than actual steady state plasma concentration values. Further, the data was from an IV study alone at one time point and did not include the subsequent oral dosage modelling data that was necessary for the inventors to translate the results of the modelling to threshold concentrations for a pulsatile release oral dosage form to be used over a 24 hour time period.

I find this passage a bit difficult to understand. The second sentence suggests that the problem is that the data is misleading. If that is the issue, Takeda was right in saying that this was a disguised s 53 attack [260]. Alternatively, Furlanetto J might be saying that the defect was simply the failure to disclosure how the inventors “arrived at the oral dosage data and steady state plasma concentrations.” This seems to be saying that the inventors must disclose in the patent the experiments and reasoning process that led them to arrive at their invention. If that is what was meant, it is not correct. As it happens, this point was addressed by Manson J in his very recent decision in Proslide v WhiteWater 2024 FC 175, in which WhiteWater argued that “the key principle” of the law of sufficiency is that “information known by the patentee may be considered in assessing whether the disclosure is sufficient,” and that “[f]acts known by the inventor, and intentionally omitted or not communicated in the description of a patent can result in disclosure being insufficient”. Manson J rejected this argument, encapsulating the established law of sufficiency as follows:

[19] WhiteWater’s position broadens the scope of sufficient disclosure beyond what the law actually supports. Section 27(3) of the Patent Act, RSC, 1985, c P-4, requires the patent to make a full disclosure, but that requirement pertains to “the invention and its operation or use as contemplated by the inventor” in the patent. Therefore, the extent of the obligation to make sufficient disclosure is limited in two ways. First, the information need only pertain to the invention as disclosed and claimed by the patent. Second, that information must enable the skilled person to make or use that invention (Teva at paras 50-52, 70). Further disclosure is not necessary to meet the requirements of sufficiency as contemplated by section 27(3).

On the whole, I suspect that what Furlanetto J was really getting at in this paragraph was similar to her second point, which is easier to understand, so I’ll turn to that.

The second point turned on the fact that the patent claims a dosage form containing an amount of PPI that results in a specified plasma concentration, but it does not disclose the amount of the PPI that is required to achieve that concentration. The amount of PPI required to achieve the specified plasma levels would depend on a variety of factors, such as the exact dosage form, as the excipients can affect absorption [263]–[264]. Consequently, determining the exact amount of PPI to achieve the specified levels would require a non-trivial amount of routine research, albeit not rising to the level of invention.

Furlanetto J cited Idenix v Gilead 2017 FCA 161 [19] for the proposition that “[a] disclosure is insufficient if it necessitates the working out of a problem,” and Seedlings 2021 FCA 154 [68], Leo Pharma 2017 FCA 50 [59] and Teva 2012 SCC 60 [75] for the proposition that “a minor research project is too much,” though some non-inventive trial and error experimentation may be permitted [257]. Applying this threshold to the facts, she held on the facts that the effort requires was too much [268].

This is problematic. Novel drugs are often patented very early in the R&D process, often on the basis of in vitro tests only, at a point where identifying the actual dose and dosage form necessary for clinical efficacy is a very long way off. This is typically true when utility is based on a sound prediction, and is often true even when utility has been demonstrated. While the patent will typically state the quantity needed, this is normally little more than guesswork. For example, in ZYTIGA 2021 FCA 45, concerning the 2,661,422 patent, which related to a combination of abiraterone acetate & prednisone for treating prostate cancer (discussed here), the specification stated that the amount to be administered was “about 0.01 mg/kg/day to about 100 mg/kg/day of abiraterone acetate” — a range of four orders of magnitude, and that for a compound that was already known. Similarly, in Rosiglitazone 2011 FC 239 (discussed here), the specification stated that the unit dose “normally contain an amount of the active ingredient in the range of from 0.1 to 1000 mg.” In HGS v Lilly [2011] UKSC 51, concerning EP(UK)0,939,804, relating to neutrokine alpha and related anitbodies, the specification stated that “[0125] As a general proposition, the total pharmaceutically effective amount of Neutrokine-α polypeptide administered parenterally per dose will be in the range of about 1 µg/kg/day to 10 mg/kg/day of patient body weight, although, as noted above, this will be subject to therapeutic discretion.”

It might be suggested that this case is different because the claims specified the requisite blood plasma concentration of the active ingredient and determining this concentration was part of the inventive concept. But sufficiency requires that the patentee disclose how to use the invention. Even when the plasma concentration is not specified, the skilled person still has to be able to practice the invention by determining an effective amount of the drug, so this doesn’t seem to be a difference in principle. A patent is sufficient if some trial and error is needed to find the clinically effective dose when a new compound is claimed, and the trial and error needed to find the dose needed to find the plasma concentration in this cases seems to be of the same nature, and if anything less extensive, given that the desired concentration is specified. But maybe there is more to this point that I am not seeing.

In coming to her conclusion, Furlanetto J relied primarily on the SCC Teva decision [267], which invalidated the patent for requiring what the decision repeatedly referred to as a “minor research project.” More specifically, and crucially, in Teva the skilled person would have to “undertake a minor research project to determine what the true invention was” [75] (my emphasis). This reference to determining the “true invention” was repeated consistently, in all the SCC’s references to a “minor research project”: see [17], [74], [75]. The issue in Teva was that the patent disclosed that one of the especially preferred compounds had been tested in humans and found to be effective, but the patent did not disclose which one. It was the failure to disclose which compound had been tested that constituted the failure to disclose “the true invention.” As the SCC noted in Teva, the trial judge felt that requiring the skilled person to undertake “a minor research project in order to determine which of the claims describes the true invention. . . plays games with the reader.” And as the FCA noted in Leo Pharma, the issue in Teva was that “the patentee had deliberately omitted essential information, thereby obscuring the fact that only one of the compounds claimed actually worked. Thus, the invention itself was not even properly disclosed” [58]. (See also Proslide [26], to the same effect.)

Thus the disclosure requirement at issue in Teva was not about disclosure of how to make the invention, it was disclosure of “the true invention.” And as the FCA pointed out in BRP v Arctic Cat 2018 FCA 172 [78], “[e]nablement (how to practice the invention) is a concept completely distinct from the disclosure of the invention itself, the latter of which was at issue in [Teva].” In this case there was no suggestion that Takeda had tried to conceal the true invention. The disclosure requirement at issue in this case was not disclosure of the true invention, but standard enablement, or how to practice the invention. And as the FCA further pointed out in Leo Pharma [58]–[59] and in BRP v Arctic Cat [78], the SCC in Teva did not change the traditional standard for sufficiency of disclosure of how to make the invention. In BRP v Arctic Cat [78], the FCA held that it was an error of law to apply the Teva standard for disclosure of the invention to the completely distinct context of enablement. With respect, it appears that Furlanetto J may have made that error in this case.

With all that said, I would return to the s 53 point. If Takeda actually knew the amount of active ingredient necessary to achieve the specified plasma concentrations and failed to disclose it, it is possible that there might be some kind of wilful misleading argument to be made. I don’t want to explore this further since it wasn’t at issue; the point is simply that a s 53 objection should be dealt with under s 53, and not by a strained interpretation of the sufficiency requirement.

Claims Broader

Apotex argued that the claims were overbroad “because the work of the inventors established that at least 1 hour was required between the pulsed doses; however, this 1 hour was not claimed” [270]. This seems to be redundant with utility, but in any event, it failed on the facts [274].

Ambiguity

A weak ambiguity attack also failed on the facts: [275]–[279].

Threshold for Sound Prediction

Sandoz Canada Inc v Janssen Inc 2023 FCA 221 Locke JA: Mactavish, Monaghan JJA affg 2022 FC 715 Pallotta J

2,659,770 / macitentan / OPSUMIT / NOC

In this brief decision, Locke JA has provided a useful summary of the law related to the threshold for a “sound” prediction of utility. The case also implicitly raises an interesting issue relating to the standard of appellate review.

Threshold for a sound prediction

At trial, Pallotta J held that Janssen’s 770 patent was valid in the face of a variety of attacks: see here. The only argument pursued on appeal related to lack of utility [2]. Pallotta J had held that while utility had not been demonstrated, it was soundly predicted. Sandoz appealed on the basis that Pallotta J has erred in law by applying too low a threshold in concluding that the prediction was “sound.” Sandoz acknowledged that Pallotta J has not explicitly misstated the law, but argued that she must have applied the wrong threshold because, according to Sandoz, the evidence could not support a finding of sound prediction based on a correct statement of the law [18]. Locke JA rejected this argument, on the basis that the threshold for sound prediction “is not high” and the facts as found by Pallotta J were sufficient to support her holding.

Locke JA’s decision provides a useful summary of the law related to the threshold for sound prediction. The leading case on sound prediction is of course Wellcome / AZT 2002 SCC 77. The leading case on the threshold is Novopharm / olanzapine 2010 FCA 197, quoted by Locke JA at [8]:

[84] [Wellcome / AZT 2002 SCC 77] does not define the threshold required for sound prediction. However, Binnie J. states that more than mere speculation is required (para. 69). He also provides the following indicia:

• the requirement is that the claims be fairly based on the patent disclosure (para. 59);

• it must be prima facie reasonable that the patentee should have a claim (para. 60);

• it cannot mean a certainty (para. 62);

• the desired result must be able to be inferred from the factual basis (para. 70).

[85] In my view, these indicia signify that a sound prediction requires a prima facie reasonable inference of utility.

The requirement of “a prima facie reasonable inference of utility” is the best known summary statement of the threshold. Pallotta J did not mention this wording explicitly, but that in itself “is not remarkable, and does not amount to an error of law” [30]. (Indeed, Sandoz itself did not explicitly refer to this wording in its written submissions before Pallotta J [31].)

Locke JA noted that Sandoz cited several additional passages from Wellcome / AZT that indicate what is not sufficient for a prediction to be sound:

1. No more than a mere belief that something might be useful (para. 25);

2. Little more than an announcement of a research project (para. 64); and

3. Only a promise that a hypothesis might later prove useful (para. 84).

Locke JA concluded:

[16] In my view, there is nothing in [Novopharm / olanzapine] that represents any kind of departure from what one would glean from a complete reading of Wellcome. While it is not necessary that the prediction be certain (see Wellcome at paragraph 62), or to a regulatory standard (see Wellcome at paragraph 63), the public is entitled to a teaching that is solid (see Wellcome at paragraph 69) and accurate and meaningful (see Wellcome at paragraph 83), and based not on speculation but exact science (see Wellcome at paragraph 84).

In affirming Pallotta J, Locke JA stated that “I am not prepared to conclude that it overlooked the guidance therein that the teaching in a patent based on a sound prediction must be solid, accurate and meaningful, and based on exact science (not speculation),” citing Wellcome / AZT [69], [83], [84]. Given his reliance on Novopharm / olanzapine, the description of the threshold for sound prediction as being as being “solid, accurate and meaningful, and based on exact science (not speculation),” is an explanation of what it means for an inference to be “a prima facie reasonable,” and is not intended as a departure from the threshold set out in Novopharm.

Standard of appellate review

Sandoz’s argument also raises an interesting issue as to how to distinguish a holding of law from a finding of fact in applying the standard of appellate review. Sandoz characterized Pallotta J’s putative error as an error of law in order to invoke a correctness standard of review [11]. Sandoz did not argue that Pallotta J made a palpable and overriding error of fact. While Sandoz acknowledged that Pallotta J did not explicitly misstate the law, it argued that Pallotta J must have failed to actually apply the correct law because the evidence could not support her finding that the prediction was sound [18]. Sandoz characterized this as an extricable error of law [11].

Sandoz’s argument was along the following lines. Suppose the law is that “Facts A, B and C are all relevant to establishing sound prediction.” If the trial judge says “C is not relevant and I have not considered it,” this is clearly an error of law. Now suppose the trial judge correctly states the law, and then says “I find that facts A, B and C are not proven,” yet nonetheless concludes that sound prediction has been established. The finding that sound prediction has been established is a mixed finding of law and fact, and so is to be reviewed on a deferential standard unless there is an extricable error of law. But the law itself was correctly stated. And it doesn’t seem to be an error of fact, because no one is disputing the factual findings that A, B and C were not proven. Sandoz’s argument was that this kind of error is an error of law, on the view that since the facts are undisputed and the result was clearly erroneous, we can infer that even though the trial judge stated the law correctly, the trial judge cannot actually have applied the correct law to the facts. Sansoz’s argument seems right in an extreme case such as I have described. The difficulty is that on that argument, the only way for an appellate court to decide whether the trial judge made an error of law is to delve into how the facts were applied to the law; and how the law was applied to the facts is a classic issue of a mixed finding of fact and law, which is reviewed on a deferential standard.

So, suppose a case involved a tort action in which the question was whether the defendant had been driving negligently, and the court stated “The law is that the defendant must act as a reasonable driver would in the circumstances. I find as a fact that the defendant was driving at 45kph in a 50kph zone, while sober, on dry roads, at noon, in heavy fog. I find that the defendant was not negligent.” If the factual findings are not disputed, this is a classic example of a question of mixed fact and law, to be reviewed on a deferential standard: Housen [2002] 2 SCR 235 [28]. Now suppose the court had instead stated “The law is that the defendant must act as a reasonable driver would in the circumstances. I find as a fact that the defendant was driving at 160kph in a 50kph zone, while drunk, on black ice, at dusk, in heavy fog. I find that the defendant was not negligent.” This must be an error of some kind; but the law is correctly stated and no one is disputing the specific factual findings. Is it an extricable error of law, as Sandoz argued in this case? It is true that an extricable error of law need not be explicit and can be inferred from the result on the facts, but an extricable error law is typically capable of being articulated as a specific legal rule, even if the trial court did not do so. For example, in Galaske v O’Donnell [1994] 1 SCR 670, discussed in Housen [31], the trial judge had found the driver of a car not negligent in failing to ensure that his young passenger wore a seat belt. This implies that the driver did not owe a duty of care to his underage passenger. The failure to identify this duty was an extricable error of law. Similarly, in Rhône (The) v Peter A.B. Widener (The) [1993] 1 SCR 497, discussed in Housen [34], the extricable error of law was failure to recognize that the key factor distinguishing directing minds of a corporation from normal employees “is the capacity to exercise decision-making authority on matters of corporate policy, rather than merely to give effect to such policy on an operational basis.” Again, the extricable error can be separately articulated as a rule of law. In my negligent driving example, there is no separate rule that can be extracted.

Now, in this case, Sandoz did suggest that there is an extricable rule, namely the threshold for sound prediction. This argument makes sense. A threshold is a point on a continuum, so there may be cases that are so far from the threshold that we can say that the trial judge must have applied the wrong threshold, even if the trial judge articulated the threshold correctly. The same can be said about my negligent driving example. So far, so good.

The problem is that even if we accept that the threshold is a matter of law, actually extracting the threshold that was implicitly applied by the trial judge may be impossible except in the easiest cases. Put another way, in Galaske and The Rhône, there was a clear rule that could be extracted and a specific error identified. When the legal rule is a threshold that turns on a holistic consideration of the facts, it will not be possible to extract the rule from its application in any case which is anywhere close to the line. So, in the tort example I gave above with the extreme set of facts, we can say holistically that there must have been an error of law, even if we cannot specifically identify it. But as we get closer to the line (what if it was dusk, the driver was going slightly over the speed limit etc), it becomes increasingly difficult to extract any legal error, even though one may have been made. Put another way, the trial judge who found the driver was not negligent on the extreme facts clearly had an erroneous conception of reasonableness in their mind. As the facts get closer to the line of ‘true’ reasonableness, the trial judge’s decision will become more and more defensible, even though the trial judge is applying exactly the same erroneous legal concept. When the facts are anywhere close to the line, an appellate court will say “We affirm on the basis that the holding that there was no negligence is a question of mixed law and fact to be reviewed on a deferential standard.” But as the facts get further from the line, it becomes clear at some point that the trial judge must have been making an error of law all along.

Maybe this is all a long-winded way of saying what the SCC said in Housen [28], that the distinction between an error of law and an error of mixed law and fact “can be difficult in practice because matters of mixed law and fact fall along a spectrum of particularity.” But it seems to me that Housen was making a somewhat different point, namely that we call something a rule of law when it is broadly applicable to a range of facts and we call a finding a question of mixed law and fact when it is unique to the circumstances of the case: “If a court were to decide that driving at a certain speed on a certain road under certain conditions was negligent, its decision would not have any great value as a precedent” [28], quoting Southam [1997] 1 SCR 748 [37]. My point is somewhat different. Even when we can state a generalizable precedent, it may be difficult to extract that rule from its application, particularly when the rule involves a threshold (or more generally, any kind of line drawn along a continuum, such as reasonableness in tort), and the facts of the case are somewhere close to the line. In that context, it makes sense to say that a rule of law of is involved, as there are cases at either extreme that we can say are clearly correctly decided one way or the other, even without disputing the facts. But it is nonetheless problematic to apply a correctness standard to a decision involving the application of that law to the facts in a case which is close to the line.

This was arguably such a case, which tested the limits of what exactly constitutes a sound prediction. How should the courts deal with such a case? Locke JA held that the threshold for establishing sound prediction is “not high” and “[t]he terms ‘prima facie’ and ‘reasonable inference’ leave considerable space for a fact-finding body in reaching its conclusion” [24]. He further noted that “[t]he fact that more experimentation was required after the rat studies did not necessarily take the utility of the invention outside the scope of a sound prediction” [25]. This “considerable space” remark, in combination with his look at the facts, strikes me as being a kind of intermediate level of review. It seems to me that Locke JA did take a look at the facts, albeit somewhat cursorily, to satisfy himself that Pallotta J had not obviously misapplied the law to the facts. That is not exactly standard doctrine on the standard of appellate review—Locke JA doesn’t seem to have been applying either an entirely deferential standard, or a correctness standard—but it strikes me as a sound approach.

Overbreadth Overlaps with Utility on the Facts

Janssen Inc v Sandoz Canada Inc 2022 FC 715 Pallotta J

2,659,770 / macitentan / OPSUMIT / NOC

In this NOC proceeding Sandoz conceded infringement and challenged the validity of the 770 patent on the basis of obviousness, lack of utility, overbreadth and insufficiency [8]. All attacks failed on the facts. One point of more general interest is that Pallotta J followed and applied the recent decision in Abiraterone 2021 FCA 45 [41] (discussed here), in holding that a statistical analysis of the results, or statistical significance, is not required to establish a reliable basis for a sound prediction [234]. Also, the overbreadth attack once again failed to gain any independent traction, as it was held on the facts to be simply a rehash of the utility argument.

There are three known biological pathways affecting pulmonary blood pressure, namely the prostacyclin, nitric oxide and endothelin pathways [119]–[122]. PDE5 inhibitors work through the nitric oxide pathway [121]. Macitentan is an endothelin receptor antagonist (ERA), which, as the name suggests, works through the endothelin pathway [3]. The 770 patent relates generally to the use of combination therapy consisting of macitentan and a PDE5 inhibitor (such as sildenafil) to treat diseases involving vasoconstriction, particularly pulmonary hypertension [6], [113], [171]. So, claim 26 is to the use of macitentan in combination with sildenafil for the treatment of pulmonary hypertension [87].

Drugs operating through all three pathways were known and the main question in respect of obviousness was whether it would have been obvious to use, or at least try, a combination therapy. Pallotta J held on the facts that it was not obvious, as the evidence relating to the efficacy of combination therapies was limited and would not have provided an acceptable level of confidence that a combination therapy would work [159], [164], [185], [186]. The desirability of using a combination of an ERA and a PDE5 inhibitor consequently was not obvious [188], and even if it was, the choice of macitentan in particular was not obvious [191], as it was “effectively unknown to the skilled person” at the relevant time [193]. Nor was the invention obvious to try; even if it was obvious to try a combination therapy (and Pallotta J did not hold that it was), it was not obvious to try macitentan in particular from amongst the known ERAs [199], [206].

The utility attack was to the effect that the data disclosed in the patent was not sufficient to establish a sound prediction of utility, and particularly that the rat studies disclosed in the patent were not probative of benefit in humans, primarily on the basis that the studies themselves were flawed, and partly on the basis that animal studies (especially when flawed) are not adequately probative of efficacy in humans [221]. Pallotta J rejected these arguments on the facts [229]–[237]. The main point of general interest is that Pallotta J noted that a statistical analysis of the results, or statistical significance, is not required to establish a reliable basis for a sound prediction [234], citing Abiraterone 2021 FCA 45 [41] (discussed here.)

While utility was soundly predicted, Pallotta J held that utility had not been demonstrated, on the basis that the asserted claims relate to use of the combination for the treatment of diseases in humans, and the rat studies would not necessarily translate to the same effect in humans. However, she acknowledged that the line between sound prediction and demonstrated utility is vague, and she stated expressly that “I do not intend to suggest that there is a bright line that requires, in every case, testing in human patients in order to support demonstrated utility for a drug intended to treat a disease in humans” [246]. Her conclusion was only that the evidence in this case was not sufficient to demonstrate utility in humans [246].

In my article Overbreadth in Canadian Patent Law: Part I (2020) 33 IPJ 21, I argued that overbreadth is almost always overlaps with some other ground of invalidity. However, in Seedlings 2021 FCA 154 (blogged here) the FCA affirmed that “overbreadth remains a proper ground of invalidity” [50]. Consequently, we are now seeing overbreadth raised regularly as a ground of invalidity. But parties attacking the validity to of a patent have struggled to find an approach to overbreadth which gives it any independent force. That pattern continues in this case, in which Pallotta J held that all the overbreadth arguments were really a restatement of utility arguments, and she dismissed them for the same reason [250]–[253]. There was one overbreadth argument which she declined to consider because it had not been properly raised [252]; this also looks like a utility argument to me.

Pallotta J dismissed the insufficiency argument very briefly [254]–[260]. Sandoz’s argument is a bit difficult to understand from Pallotta J’s brief description. It strikes me as being another rehash of the utility argument.

The underlying facts were NOT in fact disclosed

Apotex Inc v Wellcome Foundation Ltd / AZT 2002 SCC 77, aff’g [2001] 1 FC 495 (FCA), var’g T-3197-90, 79 CPR (3d) 193 (FC) Wetston J

            1,238,277

The law of utility has become much quieter since AstraZeneca 2017 SCC 36, so I’ve been spending some time reviewing the case law to see where we have ended up. One issue that remains contentious relates to the enhanced disclosure requirement that arises when utility is established on the basis of a sound prediction. In Wellcome / AZT 2002 SCC 77 at paragraph 70 the SCC remarked that:

Normally, it is sufficient if the specification provides a full, clear and exact description of the nature of the invention and the manner in which it can be practised. It is generally not necessary for an inventor to provide a theory of why the invention works. Practical readers merely want to know that it does work and how to work it. In this sort of case, however, the sound prediction is to some extent the quid pro quo the applicant offers in exchange for the patent monopoly.

This suggests that something other than the normal disclosure is required, though the SCC did not go on to explain what that might be. This passage has been interpreted as requiring a heightened disclosure requirement in the case of a sound prediction, such that the factual basis for the prediction must be disclosed, or at least referred to, in the patent itself: Raloxifene 2011 FCA 220 [46]-[47] aff’g 2010 FC 915 [116]-[17]. This enhanced disclosure requirement is substantively controversial: see Gauthier JA’s remarks in Clopidogrel 2013 FCA 186 [132] and Rennie J’s remarks in Nexium 2014 FC 638 [158-59]. Even parsing the SCC’s statement is not easy: see Rennie J’s careful discussion in Nexium [139-60] arguing that “this sort of case” refers to a new use, not sound prediction generally.

In this post, I am not going to tackle either the interpretation of paragraph 70 or the substantive merits of an enhanced disclosure requirement. Instead, I will focus on a purely factual point. The reason the SCC gave for not elaborating further is that the precise disclosure requirements “do not arise for decision in this case because both the underlying facts (the test data) and the line of reasoning (the chain terminator effect) were in fact disclosed, and disclosure in this respect did not become an issue between the parties” [70, my emphasis]. Similarly, the SCC said that “[t]he trial judge has found that the inventors possessed and disclosed in the patent both the factual data on which to base a prediction,” as well as the line of reasoning necessary to support a sound prediction [75, my emphasis]. In this post I will show that the SCC’s statement that the underlying facts supporting the sound prediction “were in fact disclosed,” is, in fact, false.

This is a purely factual point, involving a comparison between what was disclosed in the patent with the evidence that was found to support a sound prediction. The bulk of this post is devoted to supporting my claim that the factual basis was not disclosed by going through the decision and the patent in rather pedantic detail. It may seem at times that I am belabouring the obvious, but since I am arguing that the SCC made a factual error, I would like to be clear as to the basis for my assertion. I’ll finish by considering the doctrinal implications of this observation.

I should stress that I am not disagreeing with any of the key holdings in AZT. I am arguing that the SCC’s statement at [70] was based on a factually erroneous premise; but that statement was expressly obiter. Given that the remarks were obiter, it is understandable that the SCC was perhaps not as careful in dealing with the issue as it might have been. An obiter comment of this type is clearly not binding in any event: R v Henry 2005 SCC 76 [57]. My argument is that the underlying factual error is one more reason for not placing significant weight on this obiter comment in the future development of the law on this point.

Subjective or Objective Test for Sound Prediction?

Teva Canada Limited v Leo Pharma Inc 2017 FCA 50 Gauthier JA: De Montigny, Gleason JJA aff’g 2015 FC 1237 Locke J
            2,370,565 / calcipotriol & betamethasone / DOVOBET

In Teva v Leo Pharma the FCA affirmed the decision of Locke J (blogged here) granting an order of prohibition. Yesterday's post discussed the FCA holding regarding sufficiency. Teva also raised an interesting technical argument regarding sound prediction.

The test for sound prediction set out in AZT, 2002 SCC 77, requires a sound line of reasoning from which the desired result can be inferred from the factual basis [11]. In Bell Helicopter 2013 FCA 219 [154], the FCA held that the elements of the sound prediction doctrine, including the line of reasoning, need not be explicitly disclosed in the specification if they would be self-evident to the skilled person [12]. Teva argued that the line of reasoning which supports a sound prediction must nonetheless be the actual line of reasoning followed by the inventors of the invention at hand [15.1]. Teva referred to this, appropriately, as a “subjective” approach [28]. This is in contrast to an objective approach, under which it would suffice if a sound line of reasoning would be apparent to the notional skilled person, whether or not the inventors did not have that particular line of reasoning in mind.

The FCA rejected this argument, treating it essentially as a factual matter. It is undisputed that a skilled person would have to have a sound line of reasoning, and it is open to the court of first instance to infer, as a matter of fact, that the inventors shared that line of reasoning, even if the inventors were not personally called to testify [28]. Gauthier JA pointed out that such an inference must be permissible if the inventor had died before trial, and if it is permissible in that case, it must be generally permissible, as being simply a matter of how the patentee has chosen to prove a fact [29]-[30].

While that response was sufficient to dispose of the issue, it leaves open the technical question of whether the correct test for a sound line of reasoning is indeed subjective, or objective. Given the FCA’s point regarding how the facts may be proven, this will not often make any difference, but I’m not sure it will always be moot. Gauthier JA noted that

[31] From the perspective of a person skilled in the art, I see no difference in this case between an express sentence [stating the inventor’s line of reasoning] and conveying the same logic through technical information disclosed in the specification read as a whole. In both instances, the inventor conveys his or her logic to the person to whom the patent is addressed, thereby fulfilling its part of the patent bargain if this line of reasoning is held to be reasonable and sound at the relevant times. In short, in both instances, the person to whom the patent is addressed, namely, the person skilled in the art, understands the sound line of reasoning of the inventors premised on what is said in the patent.

In my view, this is a compelling argument for the objective approach. The patent bargain is an objective one; it is what the inventor actually gave the world, not what she might have thought she was giving the world. If an inventor truly believes that she was making full and sufficient disclosure, but in fact she was not, the patent will be invalid notwithstanding her subjective intent. The converse should also hold; if the inventor in fact make a disclosure that was sufficient for a skilled person, and thereby gave the invention to the world, that fulfils the bargain, even if the inventor subjectively did not understand her own invention.

Viagra Did Not Change the Test for Sufficiency

Teva Canada Limited v Leo Pharma Inc 2017 FCA 50 Gauthier JA: De Montigny, Gleason JJA aff’g 2015 FC 1237 Locke J
            2,370,565 / calcipotriol & betamethasone / DOVOBET

In Teva v Leo Pharma the FCA has affirmed the decision of Locke J (blogged here) granting an order of prohibition. Teva raised two arguments on appeal, relating to sound prediction and insufficiency. While the appeal was dismissed largely on the facts, the decision is noteworthy for the FCA’s brief but helpful discussion of Viagra / Sildenafil 2012 SCC 60, which clarifies both what Viagra did and did not hold.

Like Arctic Cat in 2017 FC 207, blogged here, Teva in this case argued that Viagra had changed the standard for sufficiency, so that now, not even a minor research project is permitted [57]. The FCA rejected this with a brief explanation of the true import of Viagra:

[58] The case before us is very different from the one that was before the Supreme Court of Canada in Sildenafil, where the patentee had deliberately omitted essential information, thereby obscuring the fact that only one of the compounds claimed actually worked (Sildenafil at paras 72, 73 and 76) . Thus, the invention itself was not even properly disclosed.

That is, Viagra concerns a situation where the patentee has deliberately obscured the nature of the invention. The FCA continued:

[59] Furthermore, the Supreme Court refers to the factual findings of the Federal Court in Sildenafil which found that in that case, one would require a minor research project to determine the true invention. This statement of Justice Lebel is, as mentioned, perfectly in line with the law as I understand it, not only in Canada but elsewhere, such as the U.K. and Europe, which recognizes that some non-inventive trial and error may be required to put a properly disclosed invention into practice. Had the Supreme Court wished to change the law on this point in Sildenafil, one would have expected a much more fulsome analysis of the issue and discussion of the relevant authorities. In the absence of such a discussion, I do not believe that the Supreme Court changed the law in Sildenafil in the manner Teva argues.

So, Viagra did not change the standard for sufficiency. I agree with Gauthier JA’s holdings on both these points.

Otherwise, Teva was largely trying to re-argue the facts, and the appeal failed for that reason [54].

(A post on the punitive damages issue in Airbus v Bell Helicopter 2017 FC 170 will be forthcoming soon, but I am taking a break from that case to clear up my blogging backlog.)

More Disagreement with Rennie J on Application of Enhanced Disclosure Requirement for Sound Prediction

Allergan Inc v Apotex Inc 2016 FC 344 Zinn J
            2,307,632 / gatifloxacin ophthalmic

Zinn J packed a number of legal issues into this short decision. As well as the issue of blinding the expert witness, discussed in yesterday’s post, his analysis of utility is also interesting.

First, like Barnes J in Lilly v Hospira 2016 FC 47 (blogged here), Zinn J disagreed with the view expressed by Rennie J in Apotex / esomeprazole 2014 FC 638 [141] that the requirement to disclose the factual basis and the line of reasoning for a sound prediction of utility is “limited to the context of ‘new use’ patents, assuming such a utility disclosure requirement exists at all” (see here). Zinn J’s reasoning echoed that of Barnes J:

[57] With the greatest of respect to the views of Justice Rennie, I am not prepared to depart from established jurisprudence directly on point from the higher courts by relying on obiter statements of a few judges in cases where the issue of utility was not fully and thoughtfully addressed. In my view, until the Federal Court of Appeal or the Supreme Court of Canada rules otherwise, Canadian jurisprudence is that, with the exception of matters of common general knowledge, the factual basis and the line of reasoning must be included in the patent.

Thus it is increasingly clear that Rennie J’s view will not carry the day at the FC level without further guidance from a higher court.*

Second, Zinn J applied the promise doctrine to hold that to hold that utility should be assessed against three promises set out in the disclosure, and not against the lower scintilla standard [48]. He rejected the argument that expressed advantages were merely goals, on the basis that the promised utility was “clearly and unequivocally described” in the patent [46]. This reaffirms that any positive statement of utility will normally be taken as a promise. Zinn J’s holding is not particularly remarkable taken in isolation, but taken cumulatively with other cases on the promise of the patent, it is now evident that the FCA’s Plavix 2013 FCA 186 decision, which seemed to signal a more restrained approach to applying the promise doctrine, has turned out to have had very little impact on the law.

On the facts of the case, Zinn J held that not all of the three promises of the ‘632 patent had been met. The ‘632 patent claimed a pharmaceutical composition of gatifloxacin and disodium edetate (EDTA). Only Claim 10, which claimed EDTA in an amount of 0.01 to 0.1 w/v%, was in issue. One of the three promises was that the addition of EDTA would prevent precipitation of gatifloxacin crystals [46]. Utility was not demonstrated, because limited testing had been done [50], and while Zinn J accepted that EDTA in the tested formulations disclosed in the patent – 0.05% and 0.1% w/v% – did in fact impact the precipitation of gatifloxacin, he found that there was no evidence that this would allow a prediction that the same result would be obtained with the lower levels of ETDA (down to 0.01%) specified in Claim 10 [60]. It’s not clear to me how to reconcile this finding with Zinn J’s finding in the context of obviousness that “a POSITA would have a fair and reasonable expectation that combining gatifloxacin with EDTA would produce an effective ophthalmic compound that would have the three advantages set out in the 632 Patent“ [39], given that the three advantages considered in the context of obviousness corresponded to the three promises of utility, including prevention of precipitation [27]. Perhaps it would have been obvious to a POSITA that levels of EDTA greater than 0.05% would have reduced precipitation, but it would not have been predictable that lower levels would have the same effect. 

*UPDATE:  I've just recalled that in Gilead v Idenix 2015 FC 1156 [380] Annis J agreed with Rennie J on this point. That means the cases are divided 2-2 at the FC level, so it is perhaps not so clear how this will play out in the Federal Court. In any event, the matter will have to be resolved by the FCA in due course.

Gilead v Idenix: Overview and Utility

Gilead Sciences, Inc v Idenix Pharmaceuticals, Inc 2015 FC 1156 Annis J
            2,490,191 2,527,657 / sofosbuvir / SOVALDI

In his first outing as a judge in a patent case, Annis J has delivered a decision of Arnoldian proportions (outstripping Arnold J’s own effort in the parallel UK case by 313 paragraphs). The case turned largely on the facts, with the bulk of the decision taken up with a meticulous review of the evidence on each point. The decision is primarily concerned with Gilead’s attack on the validity of Idenix’ 191 patent, which was held to be invalid for lack of utility and insufficiency (though Annis J did hold the 191 patent would have been infringed by Gilead’s product SOVALDI (sofosbuvir) had it been valid). Idenix’ counter-claim challenging Gilead’s 657 patent for lack of novelty and wilful misleading was dismissed. This post provides background and discusses the utility attack on the 191 patent. Legally the most interesting point is that Annis J followed Rennie J’s holding in Apotex / esomeprazole 2014 FC 638 (blogged here), that the requirement to disclose the basis for a sound prediction in the patent itself applies only to claims for a new use, and not to compound claims.

In the early 2000s both Idenix and Pharmasset (which was subsequently bought by Gilead), were working on nucleosides for the treatment of the Hepatitis C virus. Idenix discovered that some known nucleoside analogues with a 2’-C-Me/OH structure had activity in in vitro assays against the HCV class of viruses [27]. In 2002 and 2003 Idenix filed the US applications which were the priority documents for Idenix’s 191 patent [18]-[21]. Despite having only synthesized and tested compounds with a 2’-C-Me/OH structure, Idenix’ application also claimed compounds with a 2’-C-Me/F structure. CIPO objected to the application for lack of unity of invention, with the examiner having identified 15 different genera encompassed by the 23 “Forumlas” identified in the application. Indenix responded by narrowing the scope to the class of “Formula IX” compounds, claiming the 2’-C-Me/F genus. However, the disclosure was not amended [22]-[24]. (Presumably these compounds were selected because by that point – 2009 – it was apparent that the flouro compounds had the most promise.) The application was then granted as the 191 patent. The important point to take form this history is that none of the compounds claimed by the 191 patent had even been synthesized, much less tested, by Indenix as of the priority date, and they were barely discussed in the disclosure. Idenix started trying to synthesize the 2’-C-Me/F nucleosides in early 2002, but did not clearly succeed until March of 2005.

In the meantime, Pharmasset, having spotted an apparent gap in an earlier Idenix application, also set about synthesizing a 2’-C-Me/F nucleoside in early 2003 and succeeded a few months later (in May 2003) [42]-[43]. In the same month, Pharmasset filed the US application which ultimately matured to the 657 patent (now owned by Gilead after its acquisition of Pharmasset). Gilead subsequently commercialized SOVALDI, which is covered by the 657 patent. Anticipating an infringement action by Idenix based on the 191 patent, Gilead sought a declaration under s 60(1) that the 191 patent was invalid on the basis of lack of utility, overbreadth and insufficiency. Idenix counterclaimed for infringement and for a declaration that Gilead’s 657 patent is invalid.

Utility of the 191 Patent
It was common ground that utility of the patent had not been demonstrated as Idenix admitted that it did not test any compound falling within the scope of the claims of the ‘191 Patent until March 2005. Idenix therefore relied on sound prediction [221].

Annis J began the utility analysis by noting if the patent contains a promise, utility will be measured against that promise, though he noted that such a promise must be clear and unambiguous [227]. I must admit, though, that it is not entirely clear to me what standard for utility was actually applied. The parties agreed that the 191 patent contained a promise that the compounds “are useful” in the treatment HCV infections, in humans and other hosts [231], and this is evidently higher than the minimum scintilla necessary to establish utility in the absence of a promise (potential use against HCV would likely suffice). Annis J also held that the patent promised a satisfactory therapeutic index [238], and effectiveness in combination with low toxicity [240]. All of this suggests that utility was assessed against a standard which is higher than the minimum which would otherwise be necessary. But Annis J then stated explicitly that “the ‘191 Patent makes no promise of any specific result or level of treatment,” and “Gilead must therefore, prove that Idenix has not demonstrated or soundly predicted a scintilla of utility” [241].

In any event, this point was moot because Annis J concluded on the facts that there was no basis for a sound prediction of any antiviral activity in a 2’-C-Me/F nucleoside prior to its synthesis [255]; the patent would therefore lack utility whether the standard for utility was a mere scintilla or something higher.

The key question on the facts is whether the activity of the 2’-C-Me/F compounds could be predicted from the known activity of the 2’-C-Me/OH compounds [253]. The evidence was “overwhelming” that the answer was no, largely because the 2’ down position was recognized as a highly conserved and selective position, and fluoride is an unpredictable substituent particularly regarding potential toxicity [284], [376]. This case illustrates that the fact that a particular candidate compound is worth trying does not amount to a sound prediction of success [319].

More interesting in terms of the development of the law is that Annis J, following Rennie J’s holding in Apotex / esomeprazole 2014 FC 638 (blogged here), held that the requirement to disclose the factual basis and line of reasoning for sound prediction in the patent itself applies only to “new use” inventions [378]-[381]. Because the claim in the ‘191 patent was to a compound, the heightened disclosure requirement did not apply [380]. (Annis J went on to hold that if he was wrong, and there was such a requirement, the disclosure in the 191 patent was not adequate to support a sound prediction of utility [412].) The holding that there was no such disclosure requirement was strictly obiter (as was Rennie J’s), as Annis J had already determined there was no factual basis at all, disclosed or not. Nonetheless, while Annis J’s discussion was brief, he did not adopt Rennie J’s position out of comity, but because he agreed with the reasoning [380]. This adds to the weight of authority, though the point will ultimately have to be decided by the FCA.

FCA Affirms Validity of LUMIGAN RC Patent on the Facts

Apotex Inc v Allergan Inc 2015 FCA 137 Dawson JA: Webb, Boivin JA aff’g 2014 FC 567 O'Reilly J
            2,585,691 / bimatoprost / LUMIGAN RC

O’Reilly J’s holding that the ‘691 patent turned largely on the facts, and, unsurprisingly, the FCA has affirmed on the basis that O’Reilly J made no legal error and there was no palpable and overriding error in his factual findings.

O’Reilly J held, on the authority of Sanofi, 2008 SCC 61, [77], that it is sometimes necessary to look to the disclosure to determine the inventive concept of the claims. In my post on that aspect of his decision, I remarked that “O’Reilly J is clearly right on this point,” and the FCA has now affirmed that principle, on the same authority [7.1]

On appeal Apotex also argued that the utility of the invention was not soundly predicted “because the line of reasoning from which the desired result can be inferred was not explicitly disclosed in the patent” [8]. The FCA held that “[t]hose elements of the doctrine of sound prediction that would be self-evident to the skilled person need not be explicitly disclosed in the patent” [9]. This strikes me as no more than an explicit statement of what is implicit in the principle that the patent is addressed to a skilled person, and it does not represent any change in the practice in this respect. (See here for my discussion of the FC holding on this point.)

The FCA also affirmed at [10] that a genus patent does not anticipate a selection if the special advantages of the selection are not disclosed in the genus. As the FCA pointed out, this point was established in Sanofi [32].

Enhanced Disclosure Requirement Only in New Use Cases

AstraZeneca Canada Inc v Apotex Inc 2014 FC 638 Rennie J
            2,139,653 – esomeprazole – NEXIUM

As noted in yesterday’s post, in Apotex / esomeprazole Rennie J held that the effect of the SCC’s obiter remarks in Teva / sildenafil 2012 SCC 60 is to overturn previous FCA decisions holding that the factual basis for utility must be disclosed in the patent itself in all cases of sound prediction [142]. On Rennie J's view, the disclosure requirement is now “limited to the context of ‘new use’ patents, assuming such a utility disclosure requirement exists at all” [141]. This holding was expressly obiter [139], because he held that “none of the studies, disclosed or otherwise, demonstrate or soundly predict” the promised utility [140]. Nonetheless, it was a clear holding arrived at after thorough consideration.

To begin, Rennie J noted that “it is not in dispute that disclosure is not required for the demonstration of utility” [130]. In context, “disclosure” refers to disclosure in the patent of evidence demonstrating utility. The holding that such disclosure is not required is established law (see eg 2010 FCA 197 [74]; 2010 FCA 242 [80]; 2008 FCA 108, [56-64]).

He then noted that Wellcome / AZT 2002 SCC 77 [70] requires “proper disclosure” as the third component of the test for sound prediction [138], and he acknowledged that the FCA has interpreted this as requiring proper disclosure in all cases of sound prediction [142]. Rennie J began his analysis of why this is no longer good law with the Patent Act. He pointed out that utility and disclosure are treated separately, and that “there is no statutory basis for a requirement to disclose either the factual basis or the sound line of reasoning required to support a sound prediction of utility” [144].

Rennie J then provided a very careful analysis of the crucial paragraph [70] of Wellcome / AZT . That paragraph begins by saying “The doctrine of sound prediction has three components,” which is no doubt why the courts have previously understood “proper disclosure,” the third component, to apply to all cases of sound prediction. But Rennie J notes [146] that in discussing proper disclosure, the SCC begins by noting that elevated disclosure is not normally required:

Thirdly, there must be proper disclosure. Normally, it is sufficient if the specification provides a full, clear and exact description of the nature of the invention and the manner in which it can be practised. It is generally not necessary for an inventor to provide a theory of why the invention works. Practical readers merely want to know that it does work and how to work it.

The structure of this paragraph implies that normally, even in cases of sound prediction (the topic of the paragraph as a whole), the requirement of proper disclosure is satisfied by the standard disclosure requirement. Rennie J points out [147, his emphasis] that any elevated disclosure is an exception to this rule:

In this sort of case, however, the sound prediction is to some extent the quid pro quo the applicant offers in exchange for the patent monopoly.

The structure of this sentence implies that “this sort of case” is not all sound prediction cases, but only some sound prediction cases. As Rennie J says

[151] [I]t is clear from Justice Binnie’s reasoning that “this sort of case” is a subset of sound prediction cases and not a reference to all sound prediction cases. As he writes, “[i]n this sort of case, the sound prediction is to some extent the quid pro quo the applicant offers in exchange for the patent monopoly” (at para 70). By implication, there are other “sort[s] of case[s]” where the sound prediction is not the quid pro quo offered by the applicant.

This seems to be to be right, as a matter of logic and grammar. I must admit that I had always read the three sentences of paragraph [70] beginning with “Normally,” as simply describing the normal disclosure requirement that applied in cases other than sound prediction cases. I had therefore taken the key issue to be what constitutes “proper disclosure” in sound prediction cases generally. (Presumably the FCA read the paragraph the same way.) But on reflection I think that Rennie J’s reading of paragraph [70], in which “normally” describes the normal requirement for sound prediction cases, with an exception for “this sort of case,” is a better textual interpretation – though the paragraph is certainly not free of ambiguity. (And of course, on Wellcome / AZT alone, the question of what constitutes proper disclosure even in “this sort of case,” is left open.)

On Rennie J’s reading, the key question is what the SCC meant by “this sort of case.” He held that “this sort of case” means new use cases. AZT was indeed a new use case, which supports that interpretation. Rennie J advanced a purposive analysis in addition to his textual analysis (his emphasis):

[152] Second, and even more critically, limiting “this sort of case” to new use cases, rather than sound prediction cases generally, is consistent with the rationale provided by Justice Binnie. In a new use case (which AZT was), there may be an enhanced disclosure requirement because utility is the only thing being offered in exchange for the patent monopoly since the compound itself was previously disclosed.

Rennie J also appealed to the passage from Teva / sildenafil in which the SCC made some obiter remarks regarding sound prediction, including the following:

[38] As the courts below noted, all that is required to meet the utility requirement in s. 2 is that the invention described in the patent do what the patent says it will do, that is, that the promise of the invention be fulfilled

After quoting a passage from Wellcome / AZT [56] in which demonstrated utility and sound prediction are discussed together, the SCC went on to say:

[40] Nothing in this passage suggests that utility is a disclosure requirement; all it says is that "the utility required for patentability (s. 2) must, as of the priority date, either be demonstrated or be a sound prediction". Utility can be demonstrated by, for example, conducting tests, but this does not mean that there is a separate requirement for the disclosure of utility.

Rennie J interpreted this as saying that there is no heightened utility requirement for sound prediction generally, so overruling FCA decisions to the contrary. While that is a reasonable interpretation of the quoted passages, I’m not sure it was so broadly intended. The SCC began this discussion by saying “ I am of the view that this is not a case about sound prediction and that Teva's argument [that Claim 7 is invalid for insufficient disclosure of sound prediction] on this point must fail” [36]. What follows can be taken as an explanation of why Teva’s claim must fail, given that it is not a sound prediction case, in which case the remarks are directed at demonstrated utility, not sound prediction generally. This is consistent with the SCC’s concluding statement that “Since sound prediction is not an issue, the question whether there is an "enhanced" or "heightened" disclosure requirement with respect to sound predictions does not arise in this case and need not be addressed” [43]. However, as with para [70] from Wellcome / AZT this passage is ambiguous, and Rennie J may well be right.

Whether or not Teva / sildenafil overruled prior FCA authority, Rennie J has made a strong argument based on Wellcome / AZT that any enhanced disclosure requirement should apply only in new use cases, not all cases of sound prediction. (And of course, the SCC did not hold that there was any enhanced requirement in any kind of case, only that there was the possibility of some kind of enhanced disclosure in “this sort of case”.) But the question remains as to what that enhanced disclosure might be. Rennie J stated that disclosure of evidence of utility should be required in new use cases because

[152] Theoretically, without such an enhanced disclosure requirement in new use cases, a new use patent could consist of a single sentence alleging a new use and a reference to a prior patent disclosing the compound to which the use attaches. None of the research or studies supporting that new use would have to be disclosed. While new uses can be of tremendous importance (see AZT), such seemingly sparse patents would fairly raise concerns for the court when evaluating the bargain between innovators and the public.

I don’t see this as a particular cause for concern. The evidence supporting the new use would not have to be disclosed in the patent, but it will have to exist. It is clear on existing law that a claim to an entirely new compound could confine itself to a statement of its use, without supporting evidence of that use in the patent itself, though such evidence would have to exist, and I don’t see a principled distinction between those situations. Indeed, I am not sure that Rennie J himself was entirely persuaded by his own argument on this point, as he did not elaborate on the “concerns” that might be raised. It may be that he simply felt bound by prior FCA decisions to hold that there must be an elevated disclosure requirement in “this sort of case”. That suspicion is reinforced by his opening statement that the disclosure requirement is limited to the context of new use patents, “assuming such a utility disclosure requirement exists at all” [141].

With that said, Rennie J’s focus on new use patents is persuasive and thought-provoking. If we accept that enhanced disclosure arises in new use cases, and not sound prediction case generally, perhaps the requirement is simply to disclose the use, but not the supporting evidence. After all, from Consolboard, as affirmed in Teva / sildenafil, it is not generally necessary to make any disclosure of utility at all, so a requirement to disclose the use is enhanced relative to that standard.

In any event, Rennie J’s very careful analysis of the law related to proper disclosure is a welcome re-assessment of this difficult area, which he clearly sees as being problematic. I am very gratified that he quoted at [158] a concluding passage from my article, “Must the Factual Basis for Sound Prediction be Disclosed in the Patent?” (2012) 28CIPR 39, and he stated that “I generally agree with [Professor Siebrasse's] observations.” It will be interesting to see what, if anything, the FCA chooses to do with this issue on appeal. Rennie J has certainly given them a great deal to think about:

[160] In conclusion, I am of the view that there is no enhanced disclosure requirement in all sound prediction cases. Utility and disclosure are treated separately under the Patent Act, and consequently, should be treated separately in the jurisprudence as well.