Software Patent Eligibility

The Government Whittles Away at Life Sciences PatentsThe current U.S. Supreme Court has been noted for its hostility to patent holders in general, but the Supreme Court has been especially hostile to any sort of life sciences or software invention.  The Court has attempted to carve out special exceptions to the congressional mandate that patents will be granted for “any new and useful process, machine, manufacture, or composition of matter” to target inventions in software and the life sciences. However, it has done so unartfully by issuing vaguely worded decisions, leaving the regulated public in a cloud of great uncertainty. In an attempt to disperse this cloud of uncertainty, the United States Patent and Trademark Office (USPTO) has issued several lengthy statements as to how it will attempt to apply the Court’s decisions.

Up until now, the USPTO has provided guidance on software patents and patents for naturally derived products, but no guidance has been provided for the most vexing Supreme Court ban: the ban on methods of diagnosing disease. On May 2, 2016, the USPTO finally issued guidance on diagnostic methods and additional guidance on patenting naturally derived products. An especially helpful part of this guidance comes in the form of numerous specific examples of patent eligible and non-eligible life science inventions. The examples are summarized below.


One of the USPTO’s examples includes seven vaccine claims. The USPTO has taken the position that both killed and attenuated vaccines are patent eligible; the examples are for killed and attenuated virus particles, but the same logic should apply to other pathogens such as bacteria. Vaccines containing isolated antigens will also be allowed, if the vaccine contains another functionally significant ingredient, such as an adjuvant or a non-natural carrier that improves its function. However, the USPTO will not allow claims to a vaccine that is an isolated antigen in water or another carrier that does not affect the functioning of the vaccine. This is a significant departure from previous policy! Up until the present time, claims to any kind of pharmaceutical preparation comprising a natural compound (such as an antigen) and a pharmaceutically acceptable carrier have been allowed, so long as the compound is not naturally found in the carrier.

This policy shift will make it difficult to patent new vaccines based on isolated antigens. Antigen-based vaccines have been a boon to humanity in their ability to prompt highly specific immunity with fewer side effects and less cross-reactivity. However, apparently these can only be patented in the future in combination with other ingredients that modify their effects. Despite this guidance, note that one can still claim a method of vaccinating a person by administering the antigen, although the scope of protection differs from claiming the vaccine itself.

Medical Diagnosis

Since the Supreme Court’s renewed interest in patents, methods of diagnosing disease have been one of the areas of technology that have been the hardest hit. The Supreme Court’s ban on medical diagnosis was especially obtuse, and the USPTO has probably waited until this time to articulate its policy in order to let lower courts attempt to make sense of it. The USPTO has now taken the position that a method of diagnosis is not patent eligible if it includes only conventional steps such as obtaining a biological sample and testing the sample for the presence of a biomarker. However, the USPTO will allow patents on methods of detecting the biomarker without making a diagnosis — but only if the method of detecting the biomarker meets the ordinary patenting requirements of newness and inobviousness.  Consequently, testing for a previously unknown biomarker can be patented, but the discovery that a known biomarker has a new diagnostic value cannot. In addition, the use of new diagnostic reagents can be patented, such as a recombinant antibody or a xeno-antibody. A method of treating the disease state with a new treatment after diagnosis can also be patented.

These examples support the status quo ante at the USPTO that methods of diagnosing disease by testing for the presence or concentration of biomarkers are banned from patenting.

Food Additives

Example #30 is a hypothetical dietary sweetener, but it could apply to any newly discovered natural product that is useful as a food additive. As in the vaccine example, the USPTO will not allow a claim to a natural sweetener in water. A claim to the sweetener in water at a specific concentration range will not be allowed either, unless the concentration in question imparts a non-natural property to the sweetener. On the other hand, claims to the sweetener in combination with other ingredients, with which it does not occur naturally, will be allowed. It is interesting to note that the USPTO has taken the position that a qualifier such as “in a controlled release formulation” will be allowed, as the property of controlled release does not occur naturally.

Just as a vaccine cannot be defined as a newly discovered antigen, alone or in a generic carrier, a food additive cannot be defined as a newly discovered natural compound, alone or in solution. In order to obtain a patent, the invention must be defined as the additive in combination with another ingredient that changes the additive’s properties.

Genetic Screening

The Supreme Court’s first ban against patenting life science inventions was directed at synthetic DNA for genetic screening, and this remains a somewhat difficult technology to patent in the U.S. When it comes to genetic screening, the USPTO will not allow patents for the simple act of comparing a patient’s genome to another sequence because this is viewed as an “abstract mental step.” However, methods of genetic screening can be patented that are limited to specific chemical methods of genetic comparison, such as probe hybridization in combination with a specific method of detecting the hybridized probe. The USPTO will also allow methods to be patented that include highly specific and unusual chemical reactions.  It is particularly interesting to note that methods of DNA hybridization and amplification that omit any step for evaluating the meaning of the results for the patient will be allowed.

Based on this guidance, it will continue to be difficult to patent methods of genetic screening if the gene in question was already known, regardless of whether its function was understood. It would appear that one can claim methods of detecting newly discovered genes, but the value of such claims in the era of big genomics is highly questionable.

What Hasn’t Changed?

Prior guidance from the USPTO does not appear to have been disturbed by the new guidance.  Patents will still be allowed for DNA and RNA with artificial sequences (such as cDNA), natural products with additives such as preservatives (such as preserved fruit juice), chemically modified forms of natural compounds with modified properties or functions (such as botanical compounds, humanized antibodies, chimeric antibodies, and antibiotics), methods of treating disease by administering natural substances as medicines, genetically modified organisms (such as bacteria and human cell lines), fluorescently labelled natural compounds, and genetic vectors containing natural genes.

Previous guidance has also stated that patents will not normally be allowed for the purified forms of anything natural (nucleic acids, antibodies, proteins, cell lines, etc.), except if purification somehow changes its structure and function. For example, if purifying a compound that is naturally in solution results in an unnatural crystalline form of the compound, then the purified compound can be patented.  Isolated nucleic acids having natural sequences seem to be categorically excluded. Although adding a natural compound to a pharmaceutical carrier will not result in a patent, the USPTO has previously said that claiming a natural compound in a capsule is allowed, although the reason for this difference is not completely clear. Combining natural things without changing their function cannot be patented, but any change in function of such natural things in combination as compared to uncombined can be the basis for a patent. For example, a microbial consortium can be patented only if the bacteria in combination accomplish something that no species could accomplish alone.

How to Adapt?

The USPTO guidance does not have the force of law, and there is no guarantee that courts will follow it. For the most part, the guidance is clear and generally well-reasoned based on the Supreme Court’s recent rulings. Innovative life sciences companies can adapt to the new landscape by evaluating their technology early in development to determine whether versions of the invention that are allowable in the United States have any commercial value. If not, then concentrate on the needs of markets outside of the United States where by and large patents still cover useful, newly discovered natural substances and methods of medical diagnosis.

It may be possible to keep newly discovered natural products or biomarkers as trade secrets where no regulatory approval is required. Even when regulatory approval is required, instead of seeking regulatory approval for the non-patentable natural compound, consider generating a chemically modified version, maintaining the natural parent compound as a trade secret, then applying for patent protection and regulatory approval only for the modified version. Other alternative strategies are possible depending on the situation.

Business Method and Software Patent Eligibility: Australian and US Standards

This is the first of a two-part series comparing Australian and U.S. law and will focus on software and computer-related inventions.

While U.S. patent owners, applicants, and the bar have been seeking clarification from the courts and U.S. Patent & Trademark Office (USPTO) regarding patent-eligible subject matter, other nations have been having similar struggles. Two recent decisions by the Federal Court of Australia concerning the patentability of computerized business methods and biotechnology provide an opportunity to compare the standards used in Australian and U.S. law, and help shed new light on the legal tests used in the U.S.

On December 11, 2015, the Federal Court of Australia held that a business-method patent application was not eligible for protection in Commissioner of Patents v. RPL Central Pty Ltd. While the Federal Court had previously held in 2014 that business-method patents are not automatically excluded from protection, RPL represents the second invalidation of a business- method patent in as many cases by the Federal Court.

The invention in question concerned a computer program that retrieved assessable criteria published for qualification and competencies in Australian Vocational Education and Training. The computer program then generated a set of questions based on the criteria to guide individuals through the information-gathering process, storing responses and related documentation in a database for future use by a training organization related to the identified vocation in the response.

In Australian law, one qualification is that the patent must be a “manner of manufacture,” as opposed to mere schemes or plans. The “manner of manufacture” qualification is similar to the U.S. requirement of utility: The invention must belong to the “useful arts,” provide a material advantage, and be economically valuable. As the Federal Court noted in RPL, abstract ideas do not qualify as a “manner of manufacture.” Again, this is similar to U.S. law with respect to the exclusion of abstract ideas, except that in U.S. law it is not cabined solely within the utility requirement. Rather, the “abstract idea” is considered an exception to the four statutory categories of invention: products, processes, apparatuses, and compositions of matter.

RPL held that “[i]t is not a patentable invention simply to ‘put’ a business method ‘into’ a computer to implement the business method using the computer for its well-known and understood functions.” Stated another way, the computer cannot be “a mere tool in which the invention is performed,” but rather “must involve the creation of an artificial state of affairs where the computer is integral to the invention . . . .” The inventive aspect (“ingenuity” as termed by the Australian court) must be “in the way in which the computer is utilised,” not in the scheme, plan, or process that is being implemented.

At first blush, this sounds similar to the guide posts present in the U.S. Supreme Court’s decision in Alice Corp. v. CLS Bank Int’l. The Supreme Court held that “a mere instruction to ‘implement’ an abstract idea ‘on a computer’ . . . cannot impart patent eligibility.” Instead, citing prior cases, Alice held that the invention may be patent eligible where it “improve[s] the functioning of the computer itself,” or “effect[s] an improvement in any other technology or technical field.”

However, the Australian Federal Court’s focus on what is the inventive aspect of (or the “ingenuity of the inventor in,” as the Australians would say) the use of the computer emphasizes a particular point in a manner distinct from U.S. law. Despite some statements in recent cases that indicate otherwise, it remains U.S. law that a patent claim may be inventive either as to an individual element or “as an ordered combination.” In contrast, Australian law focuses on what is new—that is, what is the “artificial state of affairs”—which results specifically from the use of the computer. Moreover, that ingenious aspect may result from a number of different guideposts, such as redefining “normal” use of the computer, improving the functioning of the computer itself, using a new or unusual technical effect, the implementation of the computer itself, where prior methods had failed in doing so, or a new use for the computer beyond its commonly understood search and processing functions.

In this sense, Australian law is both narrower and broader than the scope of the U.S. test for subject matter eligibility. It is narrower in that it focuses on the specifically inventive elements in the disclosed technology rather than giving consideration to the ordered combination of steps or structures as a whole. On the other hand, the potential uses of a computer that may be considered inventive appears to be broader than those expressly stated in Alice. It may be that US law eventually recognizes the use of an “unusual technical effect” or even the implementation of a computer where it previously seemed inapplicable, but such innovations are not currently recognized by US courts and the USPTO.

Lastly, the Australian test may provide a path forward with respect to broader theoretical concerns of how patent subject matter eligibility relates to other aspects of patent law, such as anticipation, obviousness, and adequate disclosure. Placing the consideration of abstract ideas and other exceptions to patent eligible subject matter within the framework of utility (or “manner of manufacture”) seems to better locate the broad questions of innovation and usefulness within the legal framework. Rather than asking whether a process is simply the writing out of an “abstract idea” in word form (as virtually every process is in some sense), placing these exceptions within the concept of utility re-frames the question: “Is this a useful application of an abstract idea, scientific theory, or law of nature, and does it in fact build upon these ineligible discoveries in some useful way?” Perhaps by recasting eligibility within the utility framework, U.S. patent law can find a way out of the legal paradoxes that seem to arise from the current framework of precedents concerning patentable subject matter.