Ori Bergman is a J.D. candidate at Osgoode Hall Law School and is taking the Patent Law course.
The Utility requirement for granting a patent obligates the inventor to disclose an invention that works as predicted. Whether it works or not is seen as a question of fact. In submitting a claim, it is known that the specifications must not exceed the invention itself. Meaning, the claims that the examiner consider must have nothing more and nothing less than that which the inventor intended to obtain a patent for.
The doctrine of Sound Prediction, though, serves as the exception to this rule. It allows one to extrapolate the “utility of a proven invention to the utility of equivalent chemical compounds”(Lecture slides.) In other words, the inventor is claiming something that is yet to be fully tested and demonstrated to work in the way purported.
The Apotex v. Wellcome Case involved using a known drug for a novel purpose. The question arose whether that novel purpose, particularly the treatment of HIV, is patentable despite only being tested on a mouse virus in a Petri dish and HIV infected human cells in vitro, but not on actual humans with HIV. The question posed to the court included whether this novel purpose indeed has proper utility, not through direct demonstration but via the doctrine of Sound Prediction. Specifically, at the date of claim, did the drug, AZT, do what it was supposed to do, namely combat HIV? A negative answer denies that any new invention existed. This might be a difficult test given that what it sets out to do was never fully tested in the manner to which it is meant to be administered, i.e. was never tested on humans with HIV. The court describes 3 requirements that would satisfy them for the purposes of granting a patent based on a sound prediction: a) a factual basis for the prediction, for example, testing on animals or related cells that might hold predictive weight with humans, b) the inventor must, at the date of the patent application, be able to connect the facts to the desired result, i.e. show how this drug would actually treat HIV in humans and c) proper disclosure of all the facts. The court found that Wellcome successfully satisfied these 3 criteria at the date of patent application.
Consider the worst-case scenario in the above case; the patent is granted and as a result Wellcome gains a monopoly to sell the drug for the desired purpose. Naturally, Wellcome can essentially set a higher price than would be appropriate given competition. Further, research into this drug is drastically reduced, as it is essentially unusable for commercial use. Tragically, the drug is found to have toxic effects on humans with HIV. As the Supreme Court states, “the public should not be expected to pay an elevated price in exchange for speculation” (Apotex v. Wellcome at par 37.) All the more so should they not be exposed to a health hazard. This is what we must be concerned about when discussing sound predictions; that in effect, the patent granted really accomplishes very little of what was promised. The utility should not be left to be determined as a lucky guess or even through adequate research after the fact.
The court mentioned that “to have required Glaxo/Wellcome to demonstrate AZT’s efficacy through the clinical tests required by the Minister of Health for approval of a new drug for medical prescription would have been unfair to Glaxo/Wellcome.” The question is why this should be so? What about the consequences of a mistaken prediction? Do we not want the greatest amount of certainty in the invention’s utility? Consider that first person taking the first AZT pill administered. Wellcome would certainly be watching to see that AZT worked the way they hoped it would, but what if it doesn’t. The courts might reason that within the context of drugs, for it to be approved for sale, it must satisfy the standards of the Minister of Health. The focus of patent law must remain within its limits. Setting up a further hindrance and higher standard for pharmaceutical companies would effectively be performing a job that belongs to the Minister of Health and the FDA. In other words, the law of patents needs not be overly concerned with elements that are outside the realm of patents, such as health standards, but only with the ingenuity of the invention. The policy that does concern them is that the granting of patents does not become speculative as the consequences to the public would be unfair, i.e. no competition, reduced research and patents that do not do what they claim.
As a final point, one must also be attentive to a hindsight bias by the court; the drug in question reached the Supreme Court almost twenty years after it hit the market and displayed tremendous success both in treating HIV and monetarily for Wellcome.
The Doctrine of Sound Prediction represents the outer limit of the utility requirement in patent law. A proper application of this doctrine can facilitate granting a patent in a manner that protects the public while concurrently not setting the bar unreasonably high for inventors.
2 Responses
Although I agree with the court’s decision in Apotex Inc. v. Wellcome Foundation Ltd. 2002 SCC 77, and with most of the above post, I think that it is also worthwhile to consider the unique policy rationale that exists at the intersection of the doctrine of sound prediction, and potentially life-saving pharmaceuticals. In his decision, Justice Binnie paraphrases one of the inventors of the anti-viral compound, saying that “[the inventor] meant that HIV/AIDS sufferers should not be deprived of potential relief through unnecessary delay, scientific timidity, or corporate dithering” [13]. Binnie J. also acknowledges that pharmaceutical products may take years before their utility can be verified [66].
Within the context of the patenting of pharmaceuticals that literally save lives, any unnecessary or added delay in the granting of a patent will likely also delay delivery of the drug to the market—a delay that may very well cost patients their lives, particularly during “a world-wide health crisis referred to by many as an epidemic” (Binnie J., referring to AIDS, and quoting the trial judge [5]). Had the Wellcome inventors been unable to rely on the doctrine of sound prediction, they would have instead had to establish the utility of the compound by proving its efficacy, likely by conducting prolonged and extensive clinical trials. Fortunately, that was not the case, as the three requirements for sound prediction were met, and the inventors were granted a patent based on their prediction that their compound would help treat HIV/AIDS.
Given that lives may be at stake, the public interest in the early disclosure of new and useful potentially life-saving pharmaceuticals would appear to trump the public interest in avoiding granting monopoly rights in exchange for useless patents. It may be worth considering whether the doctrine of sound prediction should be applied in less of a “one size fits all” approach and more as a spectrum. Under the spectrum model, patent applications for potentially life-saving pharmaceuticals would require slightly less of a factual basis for prediction with respect to utility than inventions that are unlikely to save lives.
As a side-note, I found it slightly ironic that in this case, the generic pharmaceutical manufacturer was challenging the validity of the AZT patent on the basis of sound prediction. Had the prediction not been sound (i.e. had the drug not worked as predicted), it would not have resulted in significant monetary success, and but for that success, the drug would not have been a desirable target for litigation by the generic manufacturer.
There must be specified boundries for the doctorine/equivalence within which the claim limitation should be defined. Furthermore, based on importing information from the description on to the claims after considering the equivalence aspects within the specified boundries the claim interpretation can be limited and infringement one intends to practice should be decided
Comments are closed.