17 May 2015

Can the High Court Fix Australia’s Obviousness Problem?

StepsAustralia has an obvious problem with obviousness.  For over three decades, we have been out-of-step on inventive step.  The law of obviousness under the Patents Act 1952 was originally messed up by the High Court in the ‘3M Case’ (Minnesota Mining & Manufacturing Co v Beiersdorf (Australia) Ltd [1980] HCA 9).  There have since been three legislative attempts to fix it – with the original enactment of the Patents Act 1990, with the changes in the Patents Amendment Act 2001, and most recently with the passage of the Intellectual Property Laws Amendment (Raising the Bar) Act 2012.

I have written previously about Australia’s Four Laws of Inventive Step, and I once called the 3M Casepossibly the second worst ever patent judgment by the High Court of Australia’.  I was wrong.  In terms of its lasting negative impact on Australian patent law, 3M is clearly Number One.  The judgment I called the worst, Philips v Mirabella [1995] HCA 15 – for all its intrinsic awfulness – has been effectively quarantined, and for the most part no longer causes us too much trouble.  3M, however, lives on in the continuing failure of all attempts at legislative correction to produce laws consistent with the approach taken in the rest of the developed world.

In the latest chapter of this ongoing saga, on the 13th and 14th of May the Australian High Court heard oral arguments in the appeal by pharmaceutical company AstraZeneca AB against last year’s ruling of an expanded five-judge Full Bench of the Federal Court of Australia that its two patents relating to the cholesterol-lowering drug marketed as CRESTOR (active ingredient rosuvastatin) are invalid.  (I previously wrote about that decision in What is the Starting Point for Inventive Step?)

For truly dedicated watchers of Australian patent law, transcripts of day 1 and day 2 of the hearing are available.  Alternatively, the High Court web site hosts audio-visual recordings of the proceedings – each of day 1 and day 2 being in excess of four hours of not-always-riveting television.

It is clear from the arguments presented to the Court that Australia’s troubles with obviousness are not over.  And while the High Court now has an opportunity to make things a little better (or, at least, to avoid making them any worse again), ultimately I think that the only real solution will be to scrap our existing prescriptive approach to inventive step in favour of the kind of open enquiry into obviousness that is available under the laws of other developed nations.

What is Obviousness?

For an invention to be patentable, it must not be obvious to a person of ordinary skill in the relevant field of endeavour.  This makes perfect sense – what is obvious is not inventive, and vice versa.

Most countries have laws that say just this.  In the US, 35 USC 103 states that:

A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains.

Article 56 of the European Patent Convention provides that:

An invention shall be considered as involving an inventive step if, having regard to the state of the art, it is not obvious to a person skilled in the art.

The recent New Zealand Patents Act 2013 says, in section 7, that:

An invention, so far as claimed in a claim, involves an inventive step if it is not obvious to a person skilled in the art, having regard to any matter which forms part of the prior art base.

See how easy that is?  It can be set out in just a few lines.  Except in Australia, where the relevant provision at the time of filing of AstraZeneca’s patent applications stated that:

7 (2) For the purposes of this Act, an invention is to be taken to involve an inventive step when compared with the prior art base unless the invention would have been obvious to a person skilled in the relevant art in the light of the common general knowledge as it existed in the patent area before the priority date of the relevant claim, whether that knowledge is considered separately or together with either of the kinds of information mentioned in subsection (3), each of which must be considered separately.
7 (3) For the purposes of subsection (2), the kinds of information are:
(a) prior art information made publicly available in a single document or through doing a single act; and
(b) prior art information made publicly available in 2 or more related documents, or through doing 2 or more related acts, if the relationship between the documents or acts is such that a person skilled in the relevant art in the patent area would treat them as a single source of that information;
being information that the skilled person mentioned in subsection (2) could, before the priority date of the relevant claim, be reasonably expected to have ascertained, understood and regarded as relevant to work in the relevant art in the patent area.

The current version of section 7 of the Patents Act is simpler, and removes the requirement for information to be ‘ascertained, understood and regarded as relevant’, but it still explicitly separates ‘common general knowledge’ from ‘prior art information’ and stipulates that the skilled person must have been ‘reasonably expected to have combined’ multiple sources of information.

If you are not an Australian practitioner, I know what you are probably thinking: ‘WTF?!’  Indeed.

Blame it on 3M

As with so many other things in life, you cannot blame this mess on the sunshine, moonlight or good times – responsibility lies squarely at the feet of the High Court in the 3M Case, ably abetted by a number of subsequent parliaments that, through the normal processes of political compromise and inability to predict the consequences of their actions, failed to implement an effective ‘fix’.

The High Court had the best of intentions in the 3M Case – it was concerned to avoid the error of hindsight.  As the Court quite rightly said (at [116]), ‘[a]n allegation of want of inventive step is not made out by saying you may take one or two, or twenty-one or twenty-two, prior publications and then select from them appropriate extracts or pieces of information, which will add up to the invention claimed and so demonstrate that it was obvious.’  This is not the process by which invention occurs, or obviousness should be assessed.

However, where the courts and other decision-making bodies in other countries have sought to address the problem of hindsight knowledge head-on, the High Court in 3M effectively evaded it, by deciding that inventive step must be assessed taking into account only the stock-in-trade (i.e. the ‘common general knowledge’) of the ordinary worker in the relevant field.  Not only could you not select from twenty-one or twenty-two prior publications, but you could not even look to a single prior publication, even if it teaches 90% of the claimed invention, and even if provision of the final 10% is entirely routine and obvious, unless the information in the publication can be shown to be part of the magical body of ‘common general knowledge’, i.e. the stuff that everybody in the field has in their head or at their fingertips on a day-to-day basis.

Perhaps this did not seem as Stupid a proposition in 1980 as it does today.  But the parliament clearly found it a bit alarming, because it set about trying to legislate the 3M decision away in the Patents Act 1990.  The original version of section 7 allowed one document to be considered, in light of the common general knowledge, but only if the skilled person could reasonably be expected to have ascertained the document, understood it, and regarded it as relevant to the work in which that person was engaged. 

While this was an improvement over the 3M approach, it was not much of an improvement.  Allowing multiple documents to be combined under the 2001 amendments was a step in the right direction, and elimination of the ‘ascertained, understood and regarded as relevant’ hurdles in the Raising the Bar reforms should make things better again.

The bigger issue, however, is that any attempt to codify a test for obviousness is inherently problematic, as AstraZeneca’s CRESTOR/rosuvastatin case demonstrates.  The process of ‘invention’ is very different in the pharmaceutical field than it is in, say, the construction industry.  There is not one form of analysis that can distinguish what is ‘obvious’ from what is ‘inventive’ across all fields of human endeavour.  Furthermore, the division of background matter into ‘common general knowledge’ and ‘prior art information’ is wholly artificial, and tells us very little about what is, or is not, obvious in any given situation.

The Rosuvastatin Case

At at first instance (in Apotex Pty Ltd v AstraZeneca AB (No 4) [2013] FCA 162), a single judge of the Federal Court of Australia found that AstraZeneca’s claimed invention, of a treatment for elevated cholesterol using a relatively low dose of 5-10 mg of rosuvastatin, was obvious.  The advance was said to be important because the effective dose of this drug is well below that at which adverse side effects tend to occur.  Earlier cholesterol drugs required a period of ‘titration’ – gradually increasing doses, and taking blood tests, in order to determine an effective dosage in each patient, while trying to minimise side effects.  This is not necessary with rosuvastatin, because of its high effectiveness at relatively low doses.

AstraZeneca continues to maintain that this property of rosuvastatin was not obvious at the outset, and that its patent is therefore valid.  This may or may not be true.  It is not what concerns me here (and the High Court will decide).  My focus is on how we arrive at an assessment of obviousness.

The primary judge found the claims to be obvious in part because she took, as the starting point for the analysis, the background information disclosed in the patent specification.  The main question on appeal was whether this was the correct approach.  Inventors are often more than typically knowledgeable in their field, and may have access to insights and information that are not available to the person of more ordinary skill.

The appeals court, made up of five judges, determined that the correct approach is to determine the starting point based upon evidence of what was actually known and available to all persons skilled in the relevant field.  Notwithstanding that there was an earlier appeals court decision (Apotex Pty Ltd v Sanofi-Aventis [2009] FCAFC 134) suggesting otherwise, this seems to me to be the most logical and reasonable conclusion.  It cannot possibly be the case that the inventiveness of something is determined by how it is presented in the inventor’s patent specification, rather than upon objective evidence as to the relevant state of the art at the time.

However, on appeal the court still found the rosuvastatin low-dose claims to be obvious on the basis that there was one particular document in the prior art which – so it is said – clearly taught the skilled person that rosuvastatin was a very promising candidate for an improved cholesterol treatment, and that the relevant dosages would then be arrived at by non-inventive routine steps.

AstraZeneca’s Argument in the High Court

AstraZeneca’s submissions to the High Court reveal the difficulties that still exist with the Australian approach to obviousness.  Again, I want to emphasise that I am not looking to comment on the correctness of those arguments (Apotex and the other generic manufacturers in the case – not surprisingly – have a different view).  Rather, I am concerned that the very fact that the arguments exist, and must be made and decided before the High Court, are indicative of a problem.

According to AstraZeneca, the evidence shows that the expert witnesses in the case undertook searches to find prior art pertinent to developing an improved cholesterol treatment.  In one case, those searches resulted in the discovery of 4,500 abstracts, of which 19 appeared sufficiently promising to justify studying the full text of the corresponding articles.  Of those, two disclosed promising candidates, with one in particular suggesting that rosuvastatin might constitute the best way forward.

The Full Federal Court, on appeal, had determined that, armed with that one specific document, and reading it in light of the common general knowledge of the skilled person, the claimed invention was obvious.  On its face, this is what section 7 of the Patents Act 1990 requires, and is sufficient to establish a lack of inventive step.

AstraZeneca argues before the High Court that this approach was wrong, because it effectively ‘combines’ information from many documents in order to arrive at the solution to the problem of providing a better cholesterol-lowering drug treatment.  Only by conducting the searches, and by comparing and contrasting the different options – says AstraZeneca – is it possible to arrive at the conclusion that one particular prior art document reveals a compound that is such a promising candidate for having utility as a safe, low-dose cholesterol-lowering drug.

In effect, AstraZeneca submits that this process involves combining information from multiple sources of prior art information, which is expressly impermissible under the relevant version of section 7 of the Patents Act.  If this is correct, then AstraZeneca should prevail even if the ‘search-and-selection’ process is routine and obvious.

A Pragmatic Approach

AstraZeneca’s argument is appealing to me because, regardless of the merits of the claimed invention, it sets out an actual approach that may be taken by persons skilled in drug development when they are trying to identify a promising drug candidate. 

If the courts did not have to worry about distinguishing between ‘common general knowledge’ and ‘prior art information’, or working out whether documents would have been ‘ascertained, understood and regarded as relevant’, or whether they would be combined by the skilled person, then they could just look at this kind of evidence, and work out whether there was any particular skill, judgment or ‘inventiveness’ involved in the process of arriving at the claimed invention.

When the laws of other countries simply direct decision-makers to assess whether the skilled person would consider the claimed invention obvious, having regard to the ‘state of the art’ (or equivalent), this is what they ultimately require.  It is not a question of ‘common general knowledge’, or of ‘prior art information’ taken either alone or in combination.  It is a matter of looking at the state of the art, and the characteristics of the skilled person, and asking ‘would this person have found the claimed invention to be obvious?’ 

What becomes clear is that the separation of the state of the art into ‘common general knowledge’ and ‘prior art information’ is entirely artificial, as is the erection of statutory barriers to the identification, comprehension, application and combination of items in the prior art.  In truth, these are all simply possible facets of any inclusive evaluation of obviousness.

If AstraZeneca’s claims are ultimately confirmed to be invalid as a result of this artificial approach, it will be an exception to the more usual rule that Australia’s unique laws on inventive step has generally set the bar too low, allowing blindingly obvious ‘inventions’ to be patented on the basis that close prior art was excluded by one of the many hurdles set up by section 7 of the Patents Act.  However, denying protection to worthy inventions is no less a failure of the system than granting patents for trivial advances.

Conclusion – The Best Outcome for Australia?

I do not know which way the High Court will decide.  On balance, however, I think the best outcome for the Australian law would be for the Court to side with AstraZeneca.  What is needed is a flexible approach to assessing obviousness that takes into account the practices in the relevant field of endeavour, including the way in which the skilled person might be expected to bring together information available in the prior art.  Under the law as it applies to the rosuvastatin patents, such bringing-together of information is prohibited, which may be bad news for AstraZeneca’s generic competitors if the Court accepts that the process of comparing and selecting sources of information is a form of combination.

Under the current law, however, combining prior art information is no longer forbidden, and the requirement for the information to have been ‘ascertained, understood and regarded as relevant’ has been removed.  If the High Court mandates a flexible and pragmatic approach to evaluating the way in which prior art information is employed within the relevant technical field in any given case, this could help to bring Australian obviousness law further into line with other developed nations, despite the continuing presence of our undeniably unique statutory scheme.

Better yet, I would like to think that one day we will finally be able to put 3M behind us, and amend the law to state simply and clearly – as in the US, Europe and New Zealand – that a patentable invention must not be obvious to a skilled person, having regard to the state of the art.


Post a Comment

Copyright © 2014
Creative Commons License
The Patentology Blog by Dr Mark A Summerfield is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Australia License.