While the Full Court’s decision represents a loss for AstraZeneca, it places innovators at a significant advantage over generic companies in respect of follow-on inventions (such as formulations and dosage regimes). Where an innovator is familiar with a particular compound the innovator will effectively have a “head-start” over other companies.

The Full Court’s finding that the generics would have contravened the ‘infringement by supply’ provisions in s117 coupled with the Court’s comments that it may not have been appropriate to grant an injunction raises further uncertainty about the operation of those provisions.

An enlarged Full Federal Court (Besanko, Jessup, Foster, Nicholas and Yates JJ) in AstraZeneca AB v Apotex Pty Ltd [2014] FCAFC 99 has upheld a decision by Justice Jagot finding that two AstraZeneca patents protecting the blockbuster product CRESTOR (rosuvastatin) were invalid. The patents where challenged by three generic pharmaceutical companies: Apotex, Watson and Ascent Pharma.

The key issue in the appeal was identifying the proper “starting point” for assessing inventive step and, in particular, whether the problem set out in a patent speciation necessarily forms part of the “common general knowledge”. The Full Court overturned Justice Jagot’s finding that knowledge of rosuvastatin can be taken as the “starting point” as the common general knowledge must be assessed objectively.

Nevertheless, in the present case the Full Court upheld the findings that the patents were invalid on other grounds. This article only addresses the Full Court’s findings with respect to inventive step (Cation Patent), the priority date (Cation Patent), entitlement (Low Dose Patent) and novelty (Low Dose Patent).

This article also addresses the Full Court’s findings with respect to allegations that the generics would infringe the Low Dose Patent by supplying 20mg and 40mg tablets with the knowledge they would be split. The Full Court held that, had the patent been valid, the generics would have infringed under s117 of the Act, which prohibits supplying a product with reason to believe it will be put to an infringing use or directions to put the product to such a use. The Court commented that, in circumstances where only a small proportion of use would be infringing it may not be appropriate to grant an injunction. This raises further questions about the operation of the provision and the Court’s approach may be questioned on the basis that the Act deems the supply to be infringing. 


While the Court upheld Justice Jagot’s finding that the Cation Patent was invalid, it overturned the key finding in relation to the proper “starting point” for assessing inventive step.

At first instance, Justice Jagot found that the statement of the problem, which pre-supposed knowledge of rosuvastatin, should be taken into account as part of the common general knowledge, otherwise the invention (which claims a pharmaceutical composition) would be transformed “into something which includes the discovery of rosuvastatin”. Accordingly, Her Honour found that the starting point was the problem identified in the Cation Patent namely, how to create a composition with containing rosuvastatin with improved stability characteristics.

However, the Full Court found that the problem that is addressed in a patent specification does not necessarily form part of the “common general knowledge” against which inventive step is to be assessed. The Court cited the following key reasons in support of this conclusion.

  • It is not necessarily the case that the “starting point” outlined in the patent would have been apparent to the skilled person.
  • There is no requirement to describe the starting point in the specification.
  • Otherwise, information that was not public might be taken to form part of the common general knowledge which is contrary to the basic principles of patent law.

The Full Court’s reasoning is persuasive. Otherwise a patent claiming an invention might lack inventive step where the inventor’s understanding of the problem is disclosed, but be valid if the inventor did not describe the problem in the patent.

While the Full Court’s approach favours consistency and objectivity, it may be seen to provide a “head-start” for innovators who have particular knowledge about a substance when compared to others in the field. Indeed, it will not always be apparent to those in the relevant field that there is a particular problem that needs to be overcome.

This finding places innovators at an advantage with respect to follow-on patents where the active compound has not yet been marketed. Follow-on patents, such as patents for formulations, which may have been “obvious” to those with knowledge of the compound in question are nevertheless likely to be valid when tested against the lower level of knowledge of other skilled persons in the field.


In order to defeat the challenge to the validity of the Cation Patent, AstraZeneca also needed to persuade the Full Court to overturn Justice Jagot’s finding that the priority date of the Cation Patent was in August 2000 rather than the date the patent was amended in January 2005. AstraZeneca accepted that if the priority date was January 2005 the Cation Patent lacked novelty.

The 2005 amendments excluded from the claims certain inorganic salts (where the counter anion is a phosphate).  AstraZeneca argued that the effect of the amendments was to narrow the claims of the Cation Patent, with the result that the patent continued to enjoy the August 2000 priority date.

Justice Jagot found that the priority date was deferred to January 2005 because the amendments introduced a limitation on the claim (disclaiming compositions which used phosphate as the counter anion) which changed the nature of the invention.

The Full Court upheld this finding, holding that the patent before amendment did not disclose the invention (after amendment) because the patent before amendment positively recommended the use of the embodiment of the invention (i.e. counter anion is a phosphate) and this was excluded in 2005. 

This finding reinforces the need to give careful consideration to the consequences of amending a patent as it demonstrates that even narrowing amendments can be problematic. Equally, it highlights the need for those challenging the validity of a patent to consider the file history, including the scope of any amendments, in detail.


AstraZeneca acquired the rights to the compound rosuvastatin from the company responsible for its discovery, Shionogi & Co Ltd (Shionogi).

The generics argued that that the Low Dose Patent was invalid for lack of entitlement on the basis that the employees of Shionogi had contributed to the invention. Justice Jagot accepted this argument and the Full Court affirmed this finding.

AstraZeneca argued that while Shionogi had invented the compound rosuvastatin it had not proved a method of administration (i.e. dosage regimen). However, the Court found that Shionogi had conceived a method of administration and was entitled to the invention.

Further, the Full Court held that AstraZeneca could not rely on s22A and s138(4) of the Act which provide that a patent is not invalid merely because it is was not granted to the person entitled and limit the Court’s discretion to revoke a patent for lack of entitlement to circumstances in which it is “just and equitable to do so”. These provisions were introduced following Justice Jagot’s decision and the Full Court found that its function was limited to reviewing Her Honour’s decision.

Notwithstanding the introduction of s22A and s138(4) which clearly favour patentees in relation to this ground of invalidity, the Full Court’s finding highlights the need for patentees to carefully consider all contributions to the invention before filing a patent application. In the case of follow-on inventions it is important to consider whether inventors of earlier related inventions have made a material contribution to the invention in question.


Justice Jagot found that the Low Dose Patent was anticipated by an earlier patent (the 471 Patent) which disclosed a broad class of compounds with cholesterol lowering activity and a dosage regimen. The Full Court overturned this finding on the basis that while the 471 Patent disclosed a dosage range which was broad enough to include the regimen claimed in the Low Dose Patent, it did not specifically disclose the regimen (a starting dose of 5 to 10mg) claimed in the Low Dose Patent. 

Justice Jagot also found that the Low Dose patent was anticipated by an article which reported on the synthesis and characteristics of a number of compounds including rosuvastatin (the Watanabe article).  The Watanabe article discussed the activity of the compounds based on animal tests.

AstraZeneca argued that there was no disclosure of any dosage range in the Wantanabe article and Justice Jagot impermissibly took into account the common general knowledge regarding the potency of rosuvastatin and suitable dosage ranges. In this regard, Justice Jagot relied on evidence given by Professor O’Brien, an expert for the generics, in relation to inventive step.

The Full Court overturned Justice Jagot’s finding on the basis that Her Honour used the common general knowledge to supplement the disclosure in the Watanabe article which did not itself (expressly or impliedly) disclose any dosage regimen for the use of the rosuvastatin in humans.  The Court held that while the common general knowledge can be used to construe a prior art document for the purposes of assessing novelty it is impermissible to use the common general knowledge as a separate resource (as is permissible in the case of inventive step).

This aspect of the Court’s decision highlights the need to ensure that novelty challenges are based on the understanding of prior art by a skilled addressee in light of their acquired knowledge but not using that knowledge to “fill gaps” in the prior art. As this case demonstrates, the distinction can be elusive.


AstraZeneca argued that the supply of 20mg and 40mg dosages of rosuvastatin by the generics would infringe the Low Dose Patent under the ‘infringement by supply’ provisions in s117 of the Act. Under s117 a supply of a non-staple commercial product (rosuvastatin was found be such a product) will be an infringement where:

  1. the supplier had reason to believe it would be put to an infringing use (s117(2)(b)); or
  2. the supplier has provided instructions for the use of the product which would lead users to infringe (s117(2)(c)).

AstraZeneca argued that the generics were aware that consumers would engage in “tablet-splitting” and consequently follow the dosage regimen claimed in the Low Dose Patent.

At first instance Justice Jagot found that the generics would not infringe under s117 because there was insufficient evidence that the generics had reason to believe that consumers would engage in tablet splitting, though Her Honour accepted that there was a risk this may take place. In relation to Watson’s product information, which acknowledged that tablets could be split, Her Honour did not consider that this amounted to a sufficient instruction that would “induce” consumers for the purpose of s117(2)(c).

The Full Court overturned Justice Jagot’s findings in relation to s117. The Court held that, even though the proportion of consumers that would engage in tablet-splitting was likely to be small (only 2.75% of prescriptions were for “half a 20mg tablet”), the generics “had reason to believe” that at least some consumers would engage in that practice.

However, the Full Court cautioned that, when considering whether to grant an injunction, “some consideration of proportionality as between the extent of the infringing use that is forecast and the scope of any injunctive relief is warranted”. The Court noted that, where only a small proportion of use is likely to be infringing, it should be more difficult to obtain an injunction. Similarly, the Full Court was satisfied that Watson’s product information provided a relevant instruction for the purposes of s117(2)(c).

While the Full Court’s comments are clearly motivated by a desire to achieve a “fair” and “proportionate” outcome, the approach may be challenged on the basis that the Act deems the supply to be infringing irrespective of the end use.