One of the grounds most often claimed by parties seeking to oppose or invalidate a patent is that of ‘obviousness’ or ‘lack of inventive step’ – in summary, that the invention would have been obvious to a skilled person based on existing common knowledge and relevant publications at the time the patent was applied for. In a long-running case in Australia, AstraZeneca has attempted to defend one of its patents by arguing that strict restrictions should be applied to how a skilled person would have been able to select and use existing documents in coming up with the ‘obvious’ invention.
A full five-membered panel of the High Court has now ended this long running patent saga by dismissing all of AstraZeneca’s grounds of appeal. The decision contains lessons for any IP holder seeking to defend a patent against a claim of obviousness.
The Court’s response to the questions raised on appeal follow a sensible line which confirms that in an assessment of inventive step it is permissible for the person of skill in the art to carry out searches of existing (or ‘prior art’) documents, review and compare those documents and select the most appropriate candidate with which to proceed. To be available for consideration the selected document must be one which could have been ‘ascertained, understood and regarded as relevant’. It is at this point that this document can be considered to determine if the claimed invention would have been obvious at the priority date of the patent.
Crestor is the brand name for AstraZeneca’s cholesterol-lowering drug rosuvastatin which is a member of the statin class of drugs and which reduces cholesterol levels by inhibiting the HMG-CoA reductase enzyme involved in cholesterol production. AstraZeneca is the registered owner of a patent protecting methods of treatment of patient’s suffering from hypercholesterolemia using a single, once daily, oral dose of rosuvastatin of between 5 to 10 mg. Relevantly, rosuvastatin itself was known prior to the priority date of the patent and so the invention described in the patent hinged around the realization that it could adequately control a patient’s cholesterol levels at the cited relatively low dosage level.
The patent had originally been challenged in front of a single trial judge on a number of grounds, including obviousness. This decision was subsequently appealed to the Full Federal Court and heard by an enlarged five judge panel, where AstraZeneca lost again. The five judge panel of the Full Court was required to provide precedential weight to overcome an earlier decision of that Court on the issue of the appropriate starting point for the assessment of obviousness. That decision and its importance was discussed in our earlier article here. In essence, the Full Court found that the appropriate starting point for an assessment of obviousness is not necessarily information within the patent itself, unless this information has been shown to be part of the common general knowledge of the skilled worker, but rather a comparison between the invention as claimed and the common general knowledge in the field of the skilled worker along with any relevant prior art information.
AstraZeneca appealed that decision on a number of points, only some of which the High Court felt necessary to comment upon.
Regarding a Prior Art Document as Relevant
The relevant inventive step test for the patent in suit sets out, in essence, that (under s 7(2)) an invention will involve an inventive step unless it would have been obvious to a person of skill in the relevant art in light of the common general knowledge (CGK), whether considered separately or together with (under s 7(3)) prior art information in a single document. Importantly, that prior art information must be such that it could have been ascertained, understood and regarded as relevant to work in the relevant art (emphasis added).
Section 7(2) of the Patents Act therefore sets out that the comparison for inventive step purposes is with the CGK either separately or when combined with a single piece of prior art information. Section 7(3) then qualifies the kind of prior art information which can be used in that comparison.
Expert evidence had established that the approach taken by a person skilled in the relevant art (determined to be a medical practitioner with specialized expertise in treating hypercholesterolemia), when attempting to identify a new statin with certain improved characteristics, would be to carry out a search which, it was shown, identified a number of potentially relevant documents. An analysis of these led to identification of a smaller number of key documents of which the “Watanabe article” and “Aoki Article” were considered to be highly relevant. The Aoki Article identified a HMG-CoA reductase inhibitor termed NK-104 (i.e. a statin which is not rosuvastatin) while the Watanabe Article presented efficacy results for a number of statins, including rosuvastatin, and demonstrated rosuvastatin to be more effective and to have fewer side effects than the other statins. The Watanabe Article also noted that rosuvastatin had progressed to clinical trials.
One of AstraZeneca’s key contentions was that s 7(2) should operate such that all documents not forming part of the CGK must be considered separately in determining whether they pass the “ascertained, understood and regarded as relevant” threshold test of s 7(3) to determine if any one document is available for consideration, with the CGK, for that inventive step comparison. Effectively, AstraZeneca argued that it was not permissible, in the process of uncovering potentially relevant prior art documents, to compare multiple documents and decide which were the most relevant to then consider in the inventive step comparison test. That is, they opined that each prior art document must be considered separately not just during the inventive step assessment but also “at the anterior step of assessing whether or not any given source of information satisfies the ‘ascertained, understood and regarded as relevant’ requirement in s 7(3)”.
The High Court was unanimous in its rejection of this contention stating that “the relevance requirement is a threshold criterion for consideration, by the court, of a prior art publication in conjunction with common general knowledge for the purpose of determining obviousness. The process leading to its identification plays no role in that determination”.
This is a sensible approach by the Court since, as noted by Justice Kiefel, “a consequence of its [AstraZeneca’s] argument would appear to be that if a search for prior art information reveals more than one document, none of them could qualify as relevant to the problem if identifying a single document as relevant involved a process that required disregarding all other documents”. Any construction of the statutory test which prevents review and consideration of the ‘best candidate’ prior art document by comparison with other such documents would appear to subvert the intent of that test.
The Alternative Routes Argument
Another of AstraZeneca’s contentions upon appeal was that, if their above contention should fail, then the approach to assessment of inventive step must consider every single document identified in the search separately in the comparison under s 7(2). This would effectively mean that the skilled person would have a number of alternative routes to choose from, some of which would lead to the invention as claimed (the Watanabe Article) and others which would not (the Aoki Article identifying a different statin as a potential candidate for clinical development). AstraZeneca argued that these alternative routes resulted, in the present case, in an evidence gap as to why the route set out by the Watanabe Article should be followed in preference to that of the Aoki Article.
Once again, the High Court was unanimous in its dismissal of this ground of appeal. Firstly, one of the experts had given evidence as to why he would prefer the Watanabe Article, and hence rosuvastatin, over the NK-104 of the Aoki Article due to the greater level of efficacy and reduced side effects and also, in part, since it was already in clinical trials. This did not assist AstraZeneca in that the Court did not see any ‘gap’ in the expert evidence. Further, the Court suggested that just because there appears to be more than one avenue of exploration for the person skilled in the art does not mean that they would have to choose one over the other or, as Justices Gageler and Keane succinctly put it, that “the notional skilled addressee might be so befuddled by an embarrassment of choices as to cease pursuit of the solution”. Effectively, each document is to be considered on its merits and if one document leads to the claimed invention then that is sufficient for a finding of obviousness even if there were other documents available which do not.
Once the above issues had been addressed, and so AstraZeneca’s attempt to keep out two relevant prior art documents (the Watanabe Article and an earlier European patent ‘the 471 patent’ directed to rosuvastatin itself) had failed, the final result was inevitable. The Court felt that a person of skill in the art, in light of the CGK, and in consideration of each of those documents, separately, would be led to the use of rosuvastatin at the claimed dosage range with a reasonable expectation of success. This was because it was established that the standard medical practice of dosage titration meant that any patient prescribed a statin would be started off on the lowest dose and this dose increased over time until the desired cholesterol-lowering effect was observed without observation of unacceptable side effects. Evidence established that starting doses of 5, 10 and 20 mg were standard practice and so the route to the invention would be established by standard trial and error. The patent was therefore deemed to be invalid for lack of inventive step.
It is worth noting that the High Court, given the outcome of their obviousness finding, felt they did not need to consider further the ‘starting point’ issue which came out of the Full Court decision and was discussed in our earlier article. This means that that decision of the Full Federal Court, in relation to patents falling under the 1990 Patents Act, remains untouched affording some certainty to patentees as to the approach taken under that Act when considering inventive step.
Some interesting points were made by the High Court in relation to consideration of secondary evidence, such as commercial success, when considering inventive step questions. While rosuvastatin (Crestor) is undoubtedly highly successful as a commercial pharmaceutical, the Court noted that this success is due to its greater potency at lower doses than other statins which is a property of rosuvastatin itself. The Court pointed out that AstraZeneca did not invent rosuvastatin itself and since the claimed dosage was conventional this commercial success did not advance AstraZeneca’s case.
Finally, it is of note that, while the legislative changes brought in by the Raising the Bar Act of 2013 have dispensed with the requirement that prior art be ‘ascertained, understood and regarded as relevant’ and so the person skilled in the art is essentially deemed to be aware of all prior art in the field, a very significant number of patents and patent applications fall under the version of the Act considered by the High Court and so the learnings from this decision will continue to apply for quite some time.