A Leading Canadian IP Case: Sound Prediction

Apotex Inc. v. Wellcome Foundation Ltd., 2002 SCC 77.

In the early 1980s, AIDS (Acquired Immune Deficiency Syndrome) became one of greatest concerns in public health.

AIDS is caused by the retrovirus HIV (human immunodeficiency virus). Around 1984, scientists at Glaxo/Wellcome postulated that the propagation of retroviruses, including HIV, could be arrested by inhibiting the “reverse transcription” phase of viral replication. Specifically, the scientists considered using a “chain terminator” compound to block the virus from making a DNA copy of its RNA genome, a key step used by the virus to make a copy of itself.

After screening hundreds of potential chain terminator compounds, the Glaxo/Wellcome scientists found the compound zidovudine (AZT) produced good results as a chain terminator. Based on this testing, Glaxo/Wellcome filed a patent application in the United Kingdom on March 16, 1985, which would provide the priority basis for the application for the invention that ultimately issued as the Canadian patent.

In the early 1990s, two generic drug companies, Apotex and Novopharm, sued Glaxo/Wellcome for a declaration that the patent was invalid. Apotex and Novopharm had alleged the patent was invalid because the necessary utility or usefulness had not been established, that the claims were overbroad, and that the disclosure was misleading because it omitted naming as co-inventors scientists from the National Institutes of Health (NIH), who had assisted with the testing of AZT.

The Federal Court and Federal Court of Appeal disagreed with these allegations, finding that some claims of the patent were valid and infringed, and the matter was appealed to the Supreme Court of Canada (SCC).

The SCC unanimously found in favour of Glaxo/Wellcome on all three issues.

With respect to the first issue, Apotex and Novopharm alleged that the utility of AZT had not been established since no human testing had been performed before the patent was filed. The SCC explained that while it would have been unfair to the public to award a patent based on speculation, in this case the patentee had a “sound prediction” which fulfilled its side of the bargain.

The SCC stated that sound prediction had three components: 1) a factual basis for the prediction; 2) an articulable and “sound” line of reasoning from which the desired result can be inferred from the factual basis; and 3) proper disclosure. In this case, the Glaxo/Wellcome scientists had tested for the chain terminator effect of AZT, and had provided the data of this testing and the line of reasoning (the chain terminator effect) in the patent. As a result, Glaxo/Wellcome had shown a sound prediction of the use of AZT for treating HIV.

Regarding the second issue, the SCC found that the claims of the patent were not too broad for claiming AZT “prophylaxis” in addition to “treatment”. Under the circumstances of the case, Glaxo/Wellcome’s prediction that the “chain terminator” effect disclosed in the patent specification was sufficient to soundly predict prophylactic properties as well as its post-infection treatment application.

Finally, the SCC determined that the NIH scientists were not “co-inventors” of the patent. A “co-inventor” is someone who is responsible for the inventive concept of the invention. In this case, although the NIH scientists were extensively involved in the testing of AZT, they did not conceive of the utility of AZT for the treatment and prophylaxis of HIV. Therefore, they could not be considered co-inventors of the patent.

Nicholas Wong notes why this decision is significant and how it has trickled down through the lower courts:

In this seminal case, the SCC sought to strike a balance between the competing interests of early disclosure of new and useful inventions, even before their utility has been verified by tests, and avoiding cluttering the public domain with patents based on mere speculation.

The SCC confirmed that in order to be patentable the utility of an invention must either be actually demonstrated at the date of filing the application for the patent, or be a sound prediction based upon information and expertise available at the relevant date. While not entirely clear at the time of the decision, subsequent decisions of the courts have accepted that the “relevant date” for assessing the soundness of the prediction is the Canadian filing date.

Notably, it has been held by the courts that the factual basis and sound line of reasoning to support a sound prediction of utility must be disclosed in the patent itself. The culmination of the AZT decision and these subsequent decisions leave patentees whose inventions had not actually been demonstrated with an interesting challenge. Namely, patents drafted prior to these decisions can be subject to a legal test for sound prediction and disclosure requirements that had not been contemplated at the time the patent applications were originally drafted.

Summary by: Cheryl Cheung

Bookmark and Share

Comments are closed.