Patent applicants in the life sciences industry often find themselves under pressure to file patent applications before extensive exemplary data has been generated. In Canada, prosecuting such applications can be challenging as the Canadian Intellectual Property Office (CIPO) takes the position that the application should include working examples representative of the scope of the claims and that post-filing data cannot be relied upon to support utility. Three recent Patent Appeal Board decisions (Commissioner’s Decision Nos. 1315, 1319 and 1323), however, confirm that such challenges are not insurmountable in view of the uniquely Canadian doctrine of “sound prediction.”

Under Canadian law, if the utility of a claimed invention has not been demonstrated at the time an application is filed, then the utility must have been “soundly predicted.” For a prediction to be “sound”: (1) there must be a factual basis for the prediction; (2) the inventor must have at the date of the patent application, an articulable and “sound” line of reasoning from which the desired result can be inferred from the factual basis; and, (3) there must be proper disclosure (Apotex v. Wellcome Foundation Ltd., 2002 SCC 77).

Most recently, Decision No. 1323 (The New York Blood Center, Inc.) reviewed a Final Action issued for Canadian Patent Application No. 2,308,623, which relates to a method for making antibodies specific to a single form (Fya or Fyb) of the polymorphic “Duffy” protein. Representative claim 1 is directed to a method comprising constructing two syngeneic transgenic mice, one expressing the Fya polymorph and the second expressing the Fyb polymorph, and immunizing one of the mice with cells from the other mouse to produce an antibody specific for the polymorph expressed in the cells used for immunization. Claims 5 and 6 are product-by-process claims directed to a hybridoma cell and a monoclonal antibody, respectively, which include the steps of the method of claim 1. Claim 7 recites an alternative method, which requires constructing only one transgenic mouse and immunizing the mouse with the other Duffy protein polymorph, the origin of which is not specified.

In the Final Action, all claims were rejected under Section 84 of the Patent Rules and Subsection 27(3) of the Patent Act on the basis that they were broader in scope than the teaching in the application. The Board considered two questions in this regard: do the claims go beyond the limits of a sound prediction, and is the disclosure sufficient that the claims do not go beyond what has been described and enabled in the specification. The first question was considered to also bring into play the question of utility, even if this was not explicitly stated in the Final Action.

In assessing whether the utility of the claims was soundly predicted, the Board considered the following. Although the ‘623 application did not include any working examples of the claimed method, a working example showing a step of the method involving the construction of a transgenic mouse that expressed the Fyb polymorph in the correct (native) conformation was present. The application also described that the Duffy protein was a known and well characterized protein, and that antibodies specific to each polymorph had been made prior to the application, but only by using human sera. Attempts to make monoclonal antibodies using established animal methods had been unsuccessful. The specification further described the reasons why the inventors believed the claimed method would overcome the previously reported failures and result in antibodies specific to the Fya and Fyb polymorphs. The Board also noted that “the Examiner did not provide any evidence of inutility nor a reasoned argument as to why it is not soundly predictable that the method would produce the defined antibodies.”

Of note is the fact the Board refused to consider post-filing data provided by the Applicants demonstrating that the method worked as claimed, stating that “after-the-fact confirmation of the utility of a purported invention is not enough to support a predicted but not demonstrated utility.”

Nonetheless, with respect to claims 1, 5 and 6, the Board found that the application disclosed both a factual basis and sound line of reasoning to support a sound prediction of utility, and that no scientific rationale had been identified that would weigh against the line of reasoning. In the Board’s opinion, however, based on the teaching in the application, the prediction was sound only insofar as the Duffy protein polymorph used for immunization was presented on a murine cell membrane. The Board therefore recommended that claim 7 be amended to include this limitation.

The Board also found that the claims were sufficiently described and enabled by the specification. The Examiner had followed Decision No. 1206 (Institut Pasteur) in rejecting the claims to hybridoma cells and monoclonal antibodies on the basis that no working example of an antibody was present in the application. In Decision No. 1323, however, the Board affirmed its more recent position that making monoclonal antibodies based on a known or fully characterized protein is considered to be a matter of routine for one skilled in the art and, as such, a working example of a monoclonal antibody is no longer required.

Accordingly, with the above-noted amendment to claim 7, all claims were found allowable.

The Board’s assessment of whether the utility of the claims was soundly predicted in Decision No. 1323 followed a similar reasoning to that in Decision No. 1315 (MIT and Children’s Medical Center Association) and Decision No. 1319 (University of South Florida). In these earlier Decisions, the utility of the entire scope of the claims was found to be soundly predicted even though the working examples in each case represented only one specific embodiment of the broadest claims. In Decision No. 1319, the Board again refused to consider post-filing data presented by the Applicant.

It is noteworthy that in all three of these recent Decisions, the applications relate to well-established technologies and it remains to be seen how the Board will approach cases relating to less established technologies. Collectively, however, these recent Decisions suggest that where working examples are lacking or limited and sound prediction will be relied upon during prosecution to establish utility, then Applicants should ensure that their applications include sufficient teaching, including a description of what is already known in the art and will be relied upon in making the prediction, to establish both a factual basis and a sound line of reasoning within the specification itself. Ideally, the application should also include at least one working example that represents an embodiment that falls within the scope of the broadest claims, or a key step of a claimed method.