The Trial Division found the patent at issue to be invalid as it did not meet the test for a valid selection, and it claimed subject matter that was already the subject of a previous patent.
The Court seems to have relied on law developed from cases under the NOC Regulations in coming to its conclusion that the patent at issue is not a proper selection patent. The Court does not seem to have cited Eli Lilly v. Novopharm ( FCA 359) where it was held it is inappropriate to rely on NOC decisions to set binding precedent on controversial questions of patent law.
The Court construed the patent to be claiming a tenfold advantage of atorvastatin over the racemic mixture. The Court then reviewed the experimental data that went to support that claim and found that it was not reliable. Furthermore, the Court did not accept any evidence that atorvastatin has an unexpected or surprising inhibitory advantage over the racemic mixture. Thus, the patent did not meet a test for a valid selection.
The full text of the decision can be found at: http://decisions.fct-cf.gc.ca/en/2008/2008fc13/2008fc13.html