Pharmaceutical patent applicants are often confronted with the issue of whether to file a new drug patent application early or wait until more substantial data has been generated on a larger number of compounds. In this connection, the Federal Court of Canada recently addressed the issue of the sufficiency of a patent disclosure to support a “promised utility” for the drug rosiglitazone.

In this case1, many of the arguments, including those of insufficiency, focused on a short passage in the description of Canadian patent no. 1,328,452 (the ‘452 Patent), which reads as follows:

It has now surprisingly been discovered that certain novel substituted-thiazolidinedione derivatives show improved blood-glucose lowering activity and are therefore of potential use in the treatment and/or prophylaxis of hyperglycaemia and are of particular use in the treatment of Type II diabetes.

These compounds are also indicated to be of potential use for the treatment and/or prophylaxis of other diseases including hyperlipidaemia, hypertension, cardiovascular disease and certain eating disorders.

No data was provided in the ‘452 Patent for rosiglitazone. Experimental results were provided for other compounds in the patent. This was not a case that was argued based on sound prediction, because the patent owner asserted that, as of the filing date of the patent, it did all that was necessary to support the invention disclosed and the utility as promised in the patent. When the ‘452 Patent was challenged in Court the patent owner produced rosiglitazone test results. The data showed that a specific dose of rosiglitazone produced a 42% lower blood glucose level in the group of mice fed with feed including rosiglitazone compared with a control group.

A generic pharmaceutical company alleged invalidity of the ‘452 Patent on the basis that, among other things, there was nothing in the patent that supported the patentee’s assertion that: (1) there was a “surprising” discovery; (2) rosiglitazone allowed for an “improved” activity; and (3) rosiglitazone “will treat” type II diabetes as purportedly promised in the description of ‘452 Patent (i.e. the promised utility).

The court held that the word “improved” could be construed to mean improved compared to in the absence of a drug and the expression “of potential use” could be construed to mean that the compound may potentially be useful. More specifically, the Court held that a promise of a “potential use” of a drug was established at the time of filing, and that there is no reason for elevating “…that promise to a promise that rosiglitazone will work.” It is significant that the promise of “potential use” of a drug was sufficient to support the claims and that experimental data could be provided in court at the time the patent was challenged.

Thus, where the description promises a specific result, utility will be measured against that promise. The promise of the ‘452 Patent was construed to be that certain compounds (i.e. rosiglitazone), were “improved” in that they would lower blood glucose in obese mice compared to controls containing no compound, and that rosiglitazone had the potential to be useful. The patent did not promise that it would be clinically useful.

To avoid challenges to the validity of patents based on insufficiency of disclosure and inutility, we recommend that patent applicants include available data in patent applications rather than rely on providing it later during litigation. Applicants should also exercise caution when describing what is promised in the application regarding utility.