Is a Sale Accompanied by an Unrestricted Implied Licence to Use?

Pharmascience Inc v Janssen Inc 2024 FCA 10 Locke JA: de Montigny CJ, Goyette JA affg Janssen Inc v Pharmascience Inc 2022 FC 62 Manson J

2,655,335 / paliperidone palmitate / INVEGA SUSTENNA / NOC

This decision addresses a tricky and interesting point regarding implied licences in the context of the first prong of the Corlac 2011 FCA 228 [162] test for inducement, which requires that “the act of infringement must have been completed by the direct infringer.” Janssen’s 335 Patent relates to a dosing regimen for long-acting injectable paliperidone palmitate formulations for the treatment of schizophrenia and related disorders. It claims a dosage regimen comprising a first loading dose of 150 mg-eq of paliperidone palmitate on day 1; a second loading dose of 100 mg-eq on day 8; and subsequent maintenance doses of 75 mg-eq monthly thereafter [2]. Pharmascience proposed to sell the loading doses, but not the maintenance doses, which would be supplied by Janssen [17]. In this NOC proceeding, Pharmascience brought a motion for summary trial on the basis of non-infringement, arguing that even though the end-users, such as physicians and patients, would practice the invention, there would be no direct infringement because any end-user must be taken to have an implied licence from Janssen to use the maintenance dose for any purpose, including using it as part of the patented combination.

At first instance, Manson J held in favour of Janssen, finding that Pharmascience would induce infringement of the 335 Patent with its generic version of INVEGA SUSTENNA [3]: see here. The action proceed to trial on the patent validity issues raised by Pharmascience. Janssen again prevailed at trial, 2022 FC 1218. An appeal has been heard and is under reserve [3]. In my post on Manson J’s decision, I had noted that his decision in this case was essentially a companion case to Janssen v Apotex 2022 FC 107, which involved the same patent and the same type of generic product. However, the issues raised on appeal were quite different, as the Apotex appeal turned on the second prong of the Corlac test: see here.

This appeal therefore turned on whether the end-users, who purchased the maintenance doses from Janssen, should be taken to have an implied licence to use those maintenance doses to practice the invention in combination with the loading doses supplied by Pharmascience.

Locke JA started by summarizing the basic rule as follows: “it is clear that the sale of a patented article without restriction includes the right to use that article as the purchaser pleases,” unless a restriction is brought home to the purchaser on sale. He cited ample authority for this proposition [16], [15]. While Janssen did not dispute this rule, Janssen relied on MacLennan 2008 FCA 35, in which the patent related to “a saw tooth and tooth holder combination.” The tooth and tooth holder were separate parts, and the combination was attached to the periphery of a circular saw disc for feller heads in the forestry industry: MacLennan [4]. If the saw struck rocks while logging, individual teeth and sometimes also the holders would be broken without damaging the saw disc. The patentee sold both the tooth holder and the teeth themselves, and the defendant manufactured and sold replacement teeth to be used by the patentee’s customers with tooth holders purchased from the patentee [24]. Janssen argued that MacLennan was “on all fours” with the case at hand [25]. Pharmascience tried to distinguish MacLennan on the basis that the issue concerned the extent of the implied right of a purchaser to repair [26]. Locke JA quite rightly rejected this as being a misreading of MacLennan [26], [27].

How then can we reconcile MacLennan with the rule that the purchaser of an embodiment of the patented invention has the right to do as they please with it? Janssen argued that “the jurisprudence concerning implied license is limited to the patented article itself” [17]. Locke JA accepted this argument, saying that in MacLennan and similarly in Angelcare 2022 FC 507 (here), “the patented invention constituted a combination invention and therefore, the sale of a mere component of it was insufficient to grant the implied right to use the entire combination” [29]. He therefore held that:

[29] [T]he sale of a mere component of [a combination invention is] insufficient to grant the implied right to use the entire combination. To grant an implied licence, the sale of the entire combination had to occur, or at least, as in Slater Steel [(1968) 55 CPR 61 (Ex Ct)], the parties’ intended use of the component at the time of sale contemplated its use in the patented combination.

So, no implied licence arose in this case because Janssen sold the maintenance doses separately and this does not imply a licence to use the combination. That justifies the result in this case, as it means that the prescribers or patients who use the Pharmascience loading doses in combination with the Janssen maintenance doses are direct infringers.

But there is a problem. When does an implied licence arise? What about when Janssen sells a maintenance dose that is used in combination with a Janssen loading dose? Surely the users are not infringers. Under the first clause of Locke JA’s statement, the sale of the maintenance dose does not grant an implied licence to the use the combination, for the same reason as when the loading doses are sold by Pharmascience. Perhaps an implied licence is granted under the second clause, since it is certainly true in that case that “the parties’ intended use of the component at the time of sale contemplated its use in the patented combination.” The problem is that is also true when Janssen sells a maintenance dose that is then used in combination with a Pharmascience loading dose. If the second branch saves the user when the loading dose is sold by Janssen, it also saves the user when the loading dose is sold by Pharmascience. The only way that it is an infringement to use the Janssen maintenance doses in combination with the Pharmascience loading doses, but not with the Janssen loading doses, is if the sale by Janssen is accompanied by a restricted implied licence, which allows the use of the Janssen product only with another Janssen product.

The root of the problem is the notion that when a licence is implied, it is a licence “without restriction [that] includes the right to use that article as the purchaser pleases.” The licence that arises in MacLennan, Angelcare and in this case, is a restricted licence, to use the article in combination with another article that is also sold by the patentee. There is no way to reconcile these cases with the notion that a sale implies an unrestricted licence to use the article, no matter how the conditions of the sale itself are qualified. If the licence to use the product is unrestricted it doesn’t matter whether we say it arises when the product is sold, or only when the combination is sold, or what the intent of the parties is; the only way to get an answer that is consistent with the cases such as MacLennan is if the implied licence is restricted. Moreover, we can’t even say that the general rule is that the sale in the context of a combination invention grants a restricted implied licence to use the article in combination with other articles sold by the patentee, because in Slater Steel the patentee only sold one component of the combination (namely a power line reinforcing rod), and the other part of the combination (the power line) was supplied by third parties. In Slater Steel, the sale did convey an unrestricted implied licence.

Thus, I would suggest that the basic rule that “the sale of a patented article without restriction includes the right to use that article as the purchaser pleases,” while well supported by the authorities cited by Locke JA, is not sound as a general proposition. The sale of a patented article will often imply a licence to the use the article on the particular facts, but this is not the same as saying that it necessarily implies an unrestricted licence to use.

The difficulty arises because the proposition that the sale implies a right to use is based on logic that conflates ownership rights with patent rights. Consider the following passage from the decision of Pratte JA in Eli Lilly v Apotex (1996), 66 CPR (3d) 329, 343 which was quoted with approval by the SCC in Eli Lilly [1998] 2 SCR 129 [99]:

A patentee has a monopoly. By virtue of his patent, he has, says section 42 of the Patent Act, "the exclusive right, privilege and liberty", during the term of the patent, "of making, constructing and using the invention and selling it to others to be used". If a patentee makes a patented article, he has, in addition to his monopoly, the ownership of that article. And the ownership of a thing involves, as everybody knows, "the right to possess and use the thing, the right to its produce and accession, and the right to destroy, encumber or alienate it" [quoting Jowitt's Dictionary of English Law]. If the patentee sells the patented article that he made, he transfers the ownership of that article to the purchaser. This means that, henceforth, the patentee no longer has any right with respect to the article which now belongs to the purchaser who, as the new owner, has the exclusive right to possess, use, enjoy, destroy or alienate it. It follows that, by selling the patented article that he made, the patentee impliedly renounces, with respect to that article, to his exclusive right under the patent of using and selling the invention. After the sale, therefore, the purchaser may do what he likes with the patented article without fear of infringing his vendor's patent.

The logic is that ownership of the thing includes the right to use it, and therefore the transfer of the thing transfers to the right to use it. But ownership rights in the tangible product and patent rights relating to that product are distinct, as Pratte JA himself recognizes, and as a consequence, transfer or ownership of tangible product embodying the patent rights does not necessarily imply any transfer of patents rights.

This is illustrated by Eli Lilly itself. The SCC held that “it is clear that, in the absence of express conditions to the contrary, a purchaser of a licensed article is entitled to deal with the article as he sees fit, so long as such dealings do not infringe the rights conferred by the patent” [101] (my emphasis). The SCC accepted that “the rights of use and sale” are inherent in the unrestricted purchase of a licensed article [98], but this does not necessarily extend to the right of making. The purchaser had bought bulk product from a licensed vendor and reformulated it into final-dosage form. The SCC held that this reformulation did not constitute “making” [101], but it is clear that if it had, that would have constituted infringement. On its face, Eli Lilly holds only that a licence to use and sell are implied by an unrestricted sale, not a licence of all patent rights. Now, this is not inconsistent with the rule stated by Locke JA, which was only that the right to use is implied. The cases cited usually include the right to sell as well. But Eli Lilly shows that transfer of ownership in the tangible property embodying a patented invention is separate from the patent rights.

If the transfer of ownership in the thing does not necessarily transfer the right to make, as is clear from Eli Lilly, on what principle does it necessarily transfer the right to use? One answer is that the right to use is inherent in the right of ownership. This was the position taken by the JCPC in National Phonograph v Menck (1911) 28 RPC 229 (JCPC) 245, in which Lord Shaw began with the principle that the owner of “ordinary goods” is entitled to “use and dispose of these as he thinks fit,” so that a transfer of the right of ownership transfers that right to use and sell the goods. An implied patent licence to use and sell was seen as a way of reconciling that principle with the patentee’s patent rights: Menck 245.

But it is not strictly true that the owner of ordinary goods has the right to “use and dispose of these as he thinks fit,” because many rights and obligations distinct from the ownership right may constrain the owner’s right to use the goods they own. For example, if I own a car, I do not have the right to drive over the speed limit on the highway, and by the same token, if I sell the car to you, you do not acquire the right to exceed the speed limit. Similarly, if I own a quantity of an unpatented drug, I nonetheless do not have the right to use it for a patented purpose, unless I have a licence from the patentee. If I sell that drug to you, I clearly do not transfer the right to use it for the patented purpose, because I am not the patentee.

In response to this, it might be said that the prohibition on speeding is imposed by the state, not by me. My sale to you does not imply that no one else will interfere with your use of the car, but only that I will not interfere with your use of the car. But if I park my car in my own driveway, my sale of the car to you does not give you the right to park in my driveway, even if you might find it convenient to do so. That is because my ownership rights in the car and my rights in my land are separate, and a grant of one does not imply a grant of the other.

The general point is that the transfer of ownership rights in a tangible product good does not generally imply a licence to use those goods free of any other rights that the vendor might have that potentially restrict what the purchaser wants to do with the purchased goods. And it is clear my sale of a tangible thing that I own cannot transfer patent rights I do not own. But what if I do own the patent rights as well as the ownership rights? Is there something special about patent rights such that an unrestricted licence to use must always be implied in such a case?

Suppose P discovers a new drug, call it ‘tramifloxin,’ which P discovered is useful in treating ulcers. P gets a patent on the drug per se and successfully markets it for treating ulcers under the brand name ULCEX. P’s product patent on tramifloxin ultimately expires and P exits the market entirely, leaving it to the generics. But because P has expertise in tramifloxin, it continues to research new uses, and discovers, surprisingly and unexpectedly, that tramifloxin is also useful in treating psoriasis. P then get a second patent for the use of tramifloxin in treating psoriasis. P then successfully markets and sells tramifloxin, under the brand name PSOREX, for treating psoriasis. In that case, P’s sale of tramifloxin would not grant anyone else a licence to use tramifloxin to treat it psoriasis. But suppose P did not exit the market for ulcer treatment after the product patent expired, but instead continued to sell ULCEX. Would P’s sale of ULCEX be accompanied by an implied licence to use it to treat psoriasis? Would it matter if P sold ULCEX through an authorized generic?

What if P discovered and patented the psoriasis use before its patent on the tramifloxin per se expired, and while it was still actively marketing ULCEX. Would it matter if ULCEX was the bigger market? Would the sale of tramifloxin under the ULCEX brand imply a licence to use it to treat psoriasis? It might be said that the answer is yes, unless P gives notice that it is to be used only for treating ulcers. But that is not desirable either for P or for society. P does not want the use of ULCEX to be restricted to treating ulcers. It is in the interest of both P and society for purchasers to develop new off-label purposes, as that will expand P’s market and cure more diseases for the benefit of society. What P wants is for ULCEX not to be used to treat psoriasis. In response to that, it might be said that P should give notice on the ULCEX that it is not to be used to treat psoriasis. But how could P have given such notice before P knew that tramifloxin could be used for treating psoriasis? Does that mean that the stock of ULCEX that was in circulation before P obtained the patent can now be used to treat psoriasis? Further, if it was a third party, not P, that had obtained the psoriasis use patent, the person purchasing ULCEX from P would not get an implied licence to use it to treat psoriasis, regardless of whether P gave any kind of notice on the ULCEX. Why should P be in a worse position than that third party? Whether P holds the patent or a third party does makes no difference to the purchaser, who in either case will be is ‘surprised’ (or not) to discover that ULCEX cannot be used to treat psoriasis. Patent rights are strict liability, and the mere fact that a purchaser might not know that it does not have unrestricted rights to use has never been sufficient grounds to imply a licence. For example, if P sold the ULCEX to a intermediary subject to an express restriction that it could not be resold or used for the purpose of treating psoriasis, a purchaser from the intermediary would not acquire a right to use it for the patented use: Isler (1906) 1 Ch 605, 611 (CA). Of if P had assigned the use patent to a third party before selling the ULCEX, the purchaser would clearly not acquire a licence to use it to treat psoriasis: Betts v Willmott (1871), LR 6 Ch App 239, 244.

I am not arguing for any particular answer to these questions. My argument is only that these are difficult questions and it is not obvious that a simple rule that the sale of the product implies a licence to use the patent. Whether a licence should be implied is a question that should be answered on the facts, and not on the basis of a broad rule. These examples are somewhat artificial, but that is because it is true that a sale of a product will often imply a right to use the patented technology embodied in that product.

On that note, it might be said that the true rule is that if P sells V a product that can only be used in a way that would infringe P’s patent, then a licence to use should be implied. Most of the leading cases relied on by Locke JA were of this type — the patent at issue claimed a product, and the use of the product would necessarily infringe, so when the patentee sells that product, it would necessarily be accompanied by an implied licence to use the product. But this theory runs right into MacLennan, as the patentee in that case sold the tooth holder which could only be used in the patented combination. That is why Locke JA was forced to restrict the rule to a case in which the entire product was sold, or the entire combination was sold, or in which the patentee sold only one component but intended it to be used in the patented combination. At this point, we do not have much of a general rule left. What is the difference between stating a general rule that is tailored to a set of very specific circumstances, and simply saying that a licence will be implied when it is reasonable to do so on the facts?

This brings me back to my original point. Patent rights and rights of ownership of an embodiment of the invention are separate rights. The transfer of ownership does not necessarily imply any transfer of the patent rights. It is clear from Eli Lilly that the sale does not imply a licence to make. There is no difference in principle in respect of a right to use. It may be that it is more common that a licence to use will be implied from a sale on the facts, but that is not a matter of principle. In the context of combinations, an implied licence to use may be restricted, or unrestricted; that turns on the facts, and is not a matter principle. I suggest that the general point that emerges from all this is simply that there is no general principle that sale of a product implied a licence to use. Whether an implied licence accompanies the sale is a matter for the specific facts.

Locke JA was in a difficult position. The basic rule that “the sale of a patented article without restriction includes the right to use that article as the purchaser pleases,” is supported by ample authority. But that rule is inconsistent with MacLennan, which was clearly correctly decided. Locke JA had to find some way to restrict the scope of the general rule. The distinction he chose is intuitively appealing, but I have argued that it is ultimately unsatisfactory. With that said, I can’t think of a better approach, short of recognizing that the basic rule is itself unsound as a general proposition.