Patentable Utility and Practical Utility

Janssen Inc v Apotex Inc 2019 FC 1355 Phelan J
            2,661,422 / abiraterone acetate & prednisone / ZYTIGA / old NOC

The somewhat unusual facts in this case raise a difficult overall puzzle for which I have no good solution. Phelan J’s decision also tangentially raises several interesting points of law, though none are crucial to the outcome. This post will describe the facts and the overarching puzzle; a subsequent post will deal with the other points of law.

Abiraterone, in combination with prednisone, is approved for the treatment of prostate cancer. In general terms, the 422 patent claims abiraterone acetate in combination with prednisone for the treatment of prostate cancer. Apotex wants to sell its version of abiraterone, but because it is approved in combination with prednisone, Apotex’ product monograph will instruct that it be used in combination with prednisone. Thus, infringement by inducement was straightforward [232], [246]. In this NOC proceeding (under the old Regulations) Apotex also raised validity attacks based on obviousness, lack of utility, and lack of “patentable subject matter.” Phelan J rejected all of these attacks, and consequnetly issued the order of prohibition.

The trick in this case is that the 422 patent apparently contemplated that prednisone would be used for its anti-cancer effect; the inventors hypothesized that prednisone would reverse resistance to abiraterone and thus act synergistically with it [40]. But it seems that this hypothesis was mostly wrong. So, as Phelan J put it, “A significant problem in this case is that the anti-cancer role of [prednisone] that is claimed in the 422 Patent when used in combination with [abiraterone] appears to no longer be understood as the main role [prednisone] plays in cancer treatment. Instead, [prednisone] is now primarily understood to address the side effects caused by [abiraterone ]” [5]. Phelan J did find that there was sufficient evidence of “a scintilla” of utility of a synergistic effect to establish demonstrated utility [221]. But this is not a typical case in which the effect in question is now well established, and the issue is that there was only evidence of a scintilla at the time of filing. There was a scintilla of evidence of synergistic effect then – though barely – and there is no more evidence now than there was then. That's why I say the hypothesis was "mostly" wrong; it had enough merit to support the requisite scintilla of patentable utility, but not enough to be of practical clinical use.*

The result is that Janssen is able to prevent Apotex from marketing abiraterone in combination with prednisone to control side effects, on the basis of a mostly wrong ‘discovery’ that prednisone acts synergistically with abiraterone. This is intuitively troubling, given that the practice of prescribing abiraterone with prednisone owes nothing to Janssen’s discovery. Nonetheless, I don’t see any flaw in Phelan J’s reasoning. (I have a few quibbles that I'll discuss in my next post, but nothing that would affect the outcome.) The result stems from the fact that the standard for utility is low, so Janssen was able to demonstrate utility on the basis of a very modest effect. Normally the low standard for utility works well, as it allows the inventor to get patent protection relatively early in the product development cycle. This has practical advantages, such as enabling product development and testing, such as clinical trials, without fear of anticipating one’s own invention; and there are also theoretical advantages, in preventing a wasteful “patent race”. In the great majority of cases, the early promise either pans out after more development, and a product is brought to market, or it doesn’t pan out, the product dies, and the patent becomes irrelevant. But in this case, the early promise did not pan out, and the combination was developed for entirely different purposes. Now the patent on the defunct quasi-invention has popped up to give a monopoly over a product that owed nothing to the inventor’s contribution.

I don’t really know what to make of this. I’m inclined to think that probably this is just a case where unusual facts have given rise to a problematic result, but we should accept that no system is perfect and it is better to address the majority of cases well and accept that sometimes rules that are generally good will lead to unsatisfactory results. Or maybe I’ve overlooked some policy angle that provides a good justification for this specific result. It’s also possible that existing law really does address this problem and I (along with Apotex and Phelan J) have overlooked the answer. Or it may be that the law should be tweaked somehow to address this particular problem; but as the saying goes, hard cases make bad law, and I wouldn’t advocate any change to address these unusual facts without taking care to make sure it doesn’t do more harm than good.

*This summary is as I understand the facts. I must say that I found the facts a bit difficult to grasp, and I’m not entirely confident that I have understood the facts correctly.