From October 2016 to March 2017 the team is joined by Guest Kats Rosie Burbidge and Eibhlin Vardy, and by InternKats Verónica Rodríguez Arguijo, Tian Lu and Hayleigh Bosher.

Thursday, 1 January 2015

The New USPTO Interim Guidance on Patent Subject Matter Eligibility

The new United States Patent and Trademark Office (USPTO) Interim Guidance on Subject Matter Eligibility was published on 16 December 2014 to replace the Guidance of March 2014. The new Guidance may be found here. The accompanying ‘Nature-Based Product Examples’ can be downloaded from the same website, as can the AliceMyriad and Mayo Supreme Court decisions. The USPTO’s ‘Quick Reference Sheet’ describing the new eligibility test can be found here. The previous Guidance can be found here.    

The Guidance has the purpose of assisting USPTO staff in examination and post-grant proceedings to determine whether claimed subject matter is eligible under 35 U.S.C. 101. The Guidance has a broad remit, being relevant to computer-implemented inventions (including software and algorithms), business methods and inventions relating to natural laws and products, i.e. chemical, medical, diagnostic and biotech inventions. The Guidance is based on the three Supreme Court decisions Alice, Myriad and Mayo.

35 U.S.C. 101 simply states:

Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title’.
Leonardo told himself that the new eligibility
Guidance would not stop him inventing
However this ‘eligibility’ requirement now has the role of defining the extent to which abstract ideas, laws of nature and natural products can be monopolised by a patent. The way the eligibility requirement has developed in the Alice, Myriad and Mayo decisions and in the Guidance of March 2014 has changed the patentability landscape in the US, turning eligibility into a high hurdle that is now applicable to many more classes of inventions. The new Guidance takes into account the vociferous criticism the USPTO received in response to the March 2014 Guidance. However, whilst the eligibility test is more lenient in the new Guidance, it continues to apply to a very broad range of technical fields.

It is difficult to know which factors led to the Supreme Court to develop the eligibility test in the way that they did. However, the Alice decision may have been influenced by the perception that too many low quality patents were being granted in the software field, for example for simply performing a known business method on a computer. Such patents were being used as the basis for monetisation by patent trolls. The Myriad and Mayo decisions might have been influenced by the perception that biotech and diagnostic patents were being granted with claims that were too broad.

To recap the events that led to the new Guidance, I believe it’s worth starting at the US Supreme Court decision Bilski of June 2010. Katposts about Bilski can be found here and here. The patent in Bilski related to hedging risk in commodity trading, i.e. it was a business method. The Federal Circuit had decided in the same case that the ‘machine or transformation’ test was the only test to determine eligibility of process claims, i.e. that the process had to be tied to a particular machine or that some article had to be physically transformed in the process. That judgment had far-reaching implications for process claims in all fields, and in particular there was concern for diagnostic inventions which would usually not pass such a test. In fact the Supreme Court disagreed with the Federal Circuit, finding that the machine or transformation test was not the sole test for eligibility. However the patent was still found to relate to ineligible matter because the invention was an ‘abstract idea’. In its judgment the Supreme Court discussed the potential detrimental effects of business method patents, such as stifling innovation and competition, and these factors were seen by the Court as relevant to eligibility.

Mayo was decided in March 2012. The invention in this case was essentially a diagnostic one where measurement of the level of a metabolite in a person would determine the subsequent administration of a related drug (see Katpost here). The Supreme Court held in the first step of their analysis that the ‘correlation’ that was the basis of the invention was not eligible matter because it was a law of nature. In the second step of analysis they held that features in the claims which related to ‘determining’ the metabolite were not sufficient to transform the claim into eligible matter because they were well known. This two-part analysis became the ‘Mayo Framework’ and influenced later decisions. The Supreme Court pointed out in its decision that 35 U.S.C.101 contained the important implicit exception that ‘Laws of nature, natural phenomena, and abstract ideas were not patentable’. It also expressed its concern that the claims covered ‘all processes that make use of the correlations after measuring metabolites, including later discovered processes that measure metabolite levels in new way’. The decision clearly had very major implications for diagnostic and personalised medicine inventions, but few recognised how important the ‘Mayo framework’ would be to assessing eligibility in other fields (see Katpost here for problems with patenting personalised medicine inventions in Europe).

Newton could not believe correlations
were no longer patentable
Myriad was decided in June 2013. This was ostensibly a case about patenting human genes (even this Katpost said so). Mostly the decision was about a product claim to a human gene. That was found to be ineligible matter because the same gene occurred in nature, but non-naturally occurring cDNA sequences were held to be eligible matter. It was not clear how far the decision extended to other natural products, such as proteins, or even simple chemicals. Most believed it could not possible do so. The Guidance issued by the USPTO at the time seemed to confirm that.

However in March 2014 the USPTO issued a further Guidance based on the Mayo and Myriad decisions which truly changed the fundamentals of what was patentable in chemistry and biotech. It provided a ‘Procedure for Subject Matter Eligibility Analysis of Claims Reciting or Involving Laws of Nature, Natural Phenomena, and/or Natural Products’. The Guidance applied an eligibility test equally to product and process claims, claims relating to any naturally occurring substance, combinations of such substances (such as gunpowder), methods of treatment and methods of diagnosing. There was a strict test for any claim that ‘involved’ a ‘judicial exception’ (i.e. that related to a natural law or product). Something ‘significantly different’ had to be provided in the claim for eligibility to be attained. Twelve factors were provided which could help to determine eligibility, but from the specific examples that were given it seemed that ‘significantly different’ was a high hurdle. It required meaningful limitations to claim scope, not at a high level of generality and not by features which well known. Merely isolating a natural substance was not enough. For method of treatment claims specific dosages might need to be recited. For diagnostic inventions the claims may have to be limited to specific ways of measurement. The Guidance caused a tremendous level of uncertainty as overnight the patentability criteria changed substantially for many types of invention.

Alice was decided in June 2014 and fundamentally changed what was patentable in the field of computer-implemented inventions (see first USPTO Guidance based on the decision here and see Katpost here).  This was the Supreme Court’s opportunity to finally tell us whether software and business methods were patentable in the US. It seemed they were, but many of them were ‘abstract ideas’ which the claims failed to transform into eligible inventions. The Mayo Framework became the test that Alice used. It was clear that the way the claim was drafted would not be enough to save it, and so referring to hardware was not sufficient to become eligible. The Supreme Court explained it wanted to prevent the monopolising of abstract ideas, but did not provide guidance on how these could be defined. Whilst ‘fundamental truths’ could be abstract ideas, it seemed that so could many commercial activities.

At the time not everyone was certain that Alice would have a huge impact. However some did feel that many patents to computer-implemented inventions had now been rendered invalid. We now know that Alice has had a huge impact in litigation and in proceedings at the USPTO. It has been central to finding invalidity in over 30 District Court cases and 4 Federal Circuit decisions. At the moment it is still unclear how a computer-implemented invention would pass the Alice eligibility test. However the recent decision ib DDR Holdings v Hotels.com, in which the patent was found to relate an eligible invention may provide clues (see PatLit post here).


Plato and his friends were very pleased
with the Supreme 
Court's stance against 
monopolising abstract ideas

Returning to the new Guidance, it differs from the previous Guidance in providing a more lenient eligibility test. In the previous test any claim ‘involving’ a law of nature, natural product or abstract idea (i.e. a judicial exception) would be required to meet the ‘significantly different’ criterion. However in the new Guidance one must first look at whether the claim as whole is ‘directed’ to a judicial exception which means assessing whether it will ‘tie up’ the judicial exception. If it does not it relates to eligible matter, and there is no need to test for it being ‘significantly more’ (the new Guidance refers to ‘significantly more’ rather than ‘significantly different’ which seems of little consequence). In addition the new Guidance gives an expanded list of criteria for how a substance can be ‘markedly different’ from a natural product. This can be based on function, chemical, physical and biological properties. An isolated natural product could be eligible matter if it possessed such a difference. From the specific examples given it is clear that ‘significantly more’ will be assessed more leniently. In the example related to a method of treatment ‘applying’ a natural product to treating a disease is sufficient for it to now relate eligible matter without the need to recite specific dosages. Curiously the Guidance does not give a specific example related to diagnosis and so that remains an area in which there will continue to be much uncertainty.

For computer-implemented inventions it is unclear to this Kat whether there is much in the new Guidance that gives hope over the previous Guidance based on the Alice decision. The new Guidance has a list of economic practices and methods of organising human activities which have been deemed ‘abstract ideas’ by the Courts, and that should be helpful. However the strict interpretation the Courts have given to the Alice eligibility test probably means the USPTO has little room for being more lenient towards computer-implemented inventions.

The USPTO will accept comments on the Guidance up to 16 March 2015, and it is expected that the Guidance will change in response to the feedback which is received.

14 comments:

Anonymous said...

Computer implemented inventions was a term invented by the EPO to make software patentable, and treat some code that runs in a programmable apparatus as an "invention", while it is clearly not an "invention".

Anonymous said...

The USPTO won't be the organisation that will sort out which computer implemented inventions are patentable. It will be the Federal Circuit. However those judges each seem to have their own opinion on the issue and DDR Holdings is evidence of that. Don't expect the uncertainty to go away soon.

MaxDrei said...

Looking over to the USA from Germany, one notices how many recent CAFC decisions are a 2:1 split. The majority delivers the "right result" while the "dissent" sets out (often with great persuasive force) the other side of the argument.
I think the CAFC judges are doing this deliberately. It is all part of a plan.

Why? To foster intelligent debate about the viability of the vague litmus tests ("abstract"? What's that then?) handed down to them by the Supreme Court.

It has taken the EPO a quarter century of evolving caselaw to get Europe's law of substantive patentability harmonised. More than 20 Technical Boards of Appeal fighting it out, which legal line is best. I like to think that the USA will have arived at a similar level of legal certainty, if we give them 25 years or so to get the job done.

Anonymous said...

SO then, since 101 (and 103) was written in 1952, the 25 years you want to give them expired in 1977.

what then are the 37 additional years for...?

Anonymous said...

Has anyone noticed that a number of American commentators post-Alice talk about assessing whether there is a technical solution to a technical problem? I'm sure I've heard of that approach somewhere else....

Alex M said...

An excellent overview by Dr Ali. Practitioners should make the most of the feedback period to make their voices on this seminal document heard.

Anonymous said...

At least here in continental Europe I have never seen an idea walking the street. It appears that all ideas we have here are abstract entities. Please, could some of the learned readers try to define more precisely what an "abstract idea" is. And, since we are at it, does the idea of an "abstract idea" relate to the problem or to the solution addressed by the invention or to both.

Anonymous said...

Anonymous @ 15:15,

Even our own Supreme Court has refused to define what the word "abstract" means.

The concept (abstract or otherwise) of "void for vagueness" rings out for the scrivining of our top Court in their attempt to rewrite our law...

Anonymous said...

I think a key point was made in Mayo about patents not preventing future ways of measuring the metabolite. The Federal Circuit has to now develop an eligibility test that prevents patents stifling future innovation. I think in that respect the Supreme Court has the right idea.

Anonymous said...

An interesting point, Anonymous @ 10:00.

One thing that sticks in my craw though - the US court system (especially the Supreme Court) explicitly does not have authority to render advisory opinions.

How is ANY effort by a court - no matter how well intentioned - that seeks to "cure" a merely possible future event** NOT fatally defective along the lines of this abiding limitation on power?

**note carefully the language that the Court has used when they provide their reasoning, particularly, the use of the word "may" as opposed to the words "have" or "will" which connote actual or unavoidable harm.

Anonymous said...

In view of the comment of 12:11 I looked up Advisory Opinions and found this article: http://ir.lawnet.fordham.edu/cgi/viewcontent.cgi?article=1079&context=flr

However in response to that comment surely the Supreme Court should be trying to explain their thinking as best as possible to help the lower courts? Surely that's part of their function?

Now that eligibility is being used to limit how much of a natural law or abstract idea is 'tied up' the further development (explaining) of that practice must be a good thing?

Anonymous said...

Thank you Anonymous @ 12:40 for the interesting 1936 unnamed author piece.

Notwithstanding the view expressed there (and of which I have not fully digested), my question at 12:11 remains on the table.

Anonymous said...

It's like they are rewriting their patent law on the fly - without the benefit of the EPC as a framework/touchstone?

Sheogorath said...

In the example related to a method of treatment ‘applying’ a natural product to treating a disease is sufficient for it to now relate eligible matter without the need to recite specific dosages.
And somewhere in England, Edward Jenner is screaming, "What about the prior art I created when I invented inoculation?"

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':