PATENTS: HEALTHCARE
Navigating patent eligibility in digital healthcare
Patent applicants with digital health inventions need to remember that being useful and novel may not be enough, warn Linda Thayer, Aaron Capron and Sneha Nyshadham of Finnegan, Henderson, Farabow, Garrett & Dunner.
Digital healthcare technologies are transforming the healthcare industry and quickly changing the way healthcare is delivered. Disruptive technologies such as artificial intelligence, together with faster processing devices, have made it possible to analyse and predict risk and outcomes, sometimes in real time.
Despite these substantial benefits, some of the most innovative solutions face significant hurdles to obtaining patent protection. In two precedential opinions the US Court of Appeals for the Federal Circuit upheld the rejection of two patent applications in the field of bioinformatics as not patentable subject matter under 35 USC §101.
Bioinformatics uses a combination of computer processing and biological or genetic information to detect and diagnose medical conditions. The rejected patent applications dealt with computerised methods and systems for determining or resolving haplotype phase. Haplotype phasing is a process for determining the parent from whom alleles (ie, versions of a gene) are inherited.
In In re Bd of Trs of the Leland Stanford Junior Univ (March 11, 2021) (Stanford I), Application No. 13/445,925 (the ’925 application) recited a method for resolving haplotype phase by receiving allele data, pedigree data, transition probability data and population linkage disequilibrium data, and determining an inheritance state using Hidden Markov Modeling. A haplotype phase was determined based on the received data and the determined inheritance state.
In In re Bd of Trs of the Leland Stanford Junior Univ, No. 2020-1288 (March 25, 2021) (Stanford II), Application No. 13/486,982 (the ’982 application) claimed a computerised method for inferring haplotype phase in a collection of unrelated individuals by receiving genotype data, imputing initial haplotype phases for the individuals, then using a Hidden Markov Model to iteratively modify an imputed haplotype phase based on local recombination and mutation rates, randomly modifying and replacing an imputed haplotype phase until a final predicted haplotype phase for an individual was extracted.
Both applications extolled the novelty and importance of the inventions. According to the ‘982 application, accurately estimating haplotype phase based on genotype data obtained through sequencing an individual’s genome plays pivotal roles in population and medical genetic studies.
The inventions of the ‘925 application purportedly improved haplotype phasing techniques and promised to revolutionise personalised healthcare by tailoring risk modification, medications, and health surveillance to patients’ individual genetic backgrounds.
”The court’s precedential holdings in these two cases may surprise and concern those that see great utility and innovation in bioinformatics methods.”
Linda Thayer
Aaron Capron
PTAB conclusions
In both cases, the patent examiner rejected the claims on grounds that the claims are drawn to abstract mathematical calculations and statistical modelling and, thus, are not patent-eligible.
On review, the Patent Trial and Appeal Board (PTAB) affirmed the rejections, applying the Alice Corp Pty v CLS Bank Int’l (2014) two-step framework established by the Supreme Court for determining patent eligibility.
In both cases, the PTAB concluded that the claims recited abstract ideas that were either mathematical concepts or mental steps such as receiving or storing. On step two of the Alice analysis, the PTAB found both applications devoid of “additional limitations” that would transform the abstract ideas into patent-eligible subject matter. Although the claims may provide a valuable contribution to science, the PTAB explained that the steps of receiving data, performing calculations using that data, storing the results, and providing the results upon request using a computer do not go beyond the well-known, routine, and conventional.
In both cases, the PTAB rejected arguments that the claims recited a series of steps that resulted in improved computer functionality, finding the claims instead “use off-the-shelf computing equipment to perform an improved mathematical analysis”.
Even claims 9 and 19 of the ‘925 application, which recited a final step of “providing the drug for treatment”, were “not directed to a specific method of treatment, do not identify specific patients, recite a specific compound, prescribe particular doses, and do not identify the resulting outcome”. It held that even if the steps, taken together, resulted in a significant discovery in haplotype prediction, that alone would not establish patent eligibility.
Appeal
On appeal, the Federal Circuit reviewed the PTAB’s legal conclusions de novo, using the same two-step Alice analysis and reaching the same conclusions that the claims were directed to patent-ineligible abstract ideas. The court likened the claims to the “mathematical algorithms for performing calculations” found patent-ineligible under §101 in Parker v Flook (1978).
The court recognised that the claims may recite a “new or different use of a mathematical process” (Stanford I) or “an enhancement to the abstract mathematical calculation” (Stanford II), but concluded those do not constitute improvements to a technical process, citing Athena Diagnostics v Mayo Collaborative Servs (2019).
The Federal Circuit found nothing in either set of claims that “saved” the claims under step two of the Alice inquiry. Like the PTAB before it, the court found no practical application, no specialised computer, memory or processor, nothing that would transform the claims into the “something more” deserving of patent eligibility. It held that even if the specific ordered combination of steps in the claims resulted in a novel method that improved accuracy and speed of the computations, the innovation would only be an improvement in the mathematical analysis itself and would remain patent-ineligible.
The court’s precedential holdings in these two cases may surprise and concern those that see great utility and innovation in bioinformatics methods which, as Stanford argued, will provide “a future of personalised medicine, where genetic and biological information will be used to tailor medicine to a specific patient”.
Uncertainty about patent eligibility in this area could lead to fewer applications being filed, which could slow innovation. Stanford had until June 11, 2021, (Stanford I) and June 25, 2021, (Stanford II) to petition the Supreme Court for review.
”Applicants with inventions in bioinformatics and other digital health areas are wise to stay abreast of developments in patent eligibility.”
Future guidance
Another opportunity for clarifying guidance on §101 jurisprudence may come from the Supreme Court next autumn if it takes up review of the Federal Circuit’s decision in American Axle & Manufacturing v Neapco Holdings. On May 3, 2021, the Supreme Court asked the Solicitor General to weigh in as to whether it should grant certiorari in the first case on patent eligibility since Alice.
While American Axle is a mechanical engineering case, the Supreme Court could use it as a vehicle to fine-tune the Alice two-step method and bridge the divide in the Federal Circuit on how to apply Alice in many situations.
Until then, applicants with inventions in bioinformatics and other digital health areas are wise to stay abreast of developments in patent eligibility and remember that having a novel and useful invention may not be enough.
In the words of the court in Stanford II: “Such basic tools are not patent-eligible. Nor is novelty the touchstone of patent eligibility.”
Linda Thayer is a partner at Finnegan, Henderson, Farabow, Garrett & Dunner. She can be contacted at: linda.thayer@finnegan.com
Aaron Capron is a partner at Finnegan, Henderson, Farabow, Garrett & Dunner. He can be contacted at: aaron.capron@finnegan.com
Sneha Nyshadham is a law clerk at Finnegan, Henderson, Farabow, Garrett & Dunner. She can be contacted at: sneha.nyshadham@finnegan.com
Images: Shutterstock / Alexander Limbach
