The Regents of the University of California v. The Broad Institute, Inc., Appeal Nos. 2022-1594, -1653 (Fed. Cir. May 12, 2025)

Must an inventor know their invention will work to demonstrate that they “conceived” of it?  In this week’s Case of the Week, the Federal Circuit says no—conception does not require that an inventor know with certainty their idea will work.  This appeal considers the PTAB’s conclusions of Interference No. 106,115 over claim 18 of U.S. Patent No. 8,697,359 (corresponding to U.S. Patent Application No. 15/981,807).  After the PTAB concluded The Broad Institute et al. (Broad) won priority to the invention at hand, Regents of the University of California et al. (Regents) appealed.  (Broad) cross-appealed.  In a very fact-heavy opinion, the Federal Circuit ultimately remanded the case to the PTAB in favor of Regents, and dismissed Broad’s cross-appeal.

The main question in the relevant interference proceeding was which party first invented the use of single-guide RNA in the CRISPR-Cas9 system to genetically edit eukaryotic cells.  The use of the single-guide RNA simplified the gene editing system, which previously required two different RNA pieces.  And, while CRISPR-Cas9 had been used in prokaryotes (e.g., bacterial cells) prior to the filing of the patent applications at hand, the application of the gene editing in eukaryotic cells (e.g., mammalian cells) had not.

Throughout 2012, both parties were actively conducting research on the use of single-guide RNA to edit eukaryotic cells using the CRISPR-Cas9 system.  While Broad appeared to be the first to actually show successful results of editing a eukaryotic cell with the system, Regents provided evidence to support that they had the plans and understanding of how to do so, before Broad demonstrated the actual reduction to practice.  Regents also argued that Broad had derived the information from them to learn how to actually reduce the invention to practice.

Going into the interference, both parties filed preliminary motions.  Broad filed two—one related to claim construction and the other related to the removal of certain claims from the interference proceeding.  The claim construction issue related to the term “guide RNA.”  Broad argued that the term was generic and covered both the single- or dual- molecule RNA, in contrast to the PTAB’s conclusion that it only encompassed the single-molecule RNA configuration.

Regent’s preliminary motion related to its entitlement of the priority date of several provisional applications filed throughout 2012.  Regents first asked for the benefit of their May 2012 filing date of their first provisional application, or for the benefit of their October 2012 filing date of their second provisional application.  If neither of those were granted, Regents subsequently requested the priority benefit of their third provisional application with a filing date of January 2013.  The PTAB determined that neither the first or second applications showed a constructive reduction to practice the use of single-guide RNA in CRISPR-Cas9 to edit eukaryotes, and therefore failed to meet written description.  Thus, the PTAB ultimately gave Regents the priority benefit of the third provisional application, noting that it did satisfy written description.  The PTAB also rejected Regents’ argument that they were entitled to a conception date of March 1, 2012 and that Broad derived the invention from Regents.

Because Broad had evidence that they achieved actual reduction practice before Regents, the PTAB concluded that Broad had priority over Regents, and that Broad was first to invent.  This decision was based on Broad reducing the invention to practice on October 5, 2012 in light of their manuscript submission to academic journal Science showing actual reduction to practice.

Regents appealed based on the findings regarding inventorship and written description of the first and second provisional applications.  Broad cross-appealed based on the claim construction issue.

Conception and inventorship are ultimately questions of law reviewed de novo.  The Federal Circuit pointed out that conception is “the formation in the mind of the inventor, of a definite and permanent idea of the complete and operative invention, as it is hereafter to be applied in practice” and “is complete only when the idea is so clearly defined in the inventor’s mind that only ordinary skill would be necessary to reduce the invention to practice, without extensive research or experimentation.”  Therefore, the Federal Circuit held that the PTAB erred in its analysis of the interference proceeding by improperly conflating the standards for conception with reduction to practice.

Ultimately, the Federal Circuit clarified that conception does not require that the inventor knows their invention will work.  Knowing that the invention will work is necessarily a part of the actual reduction to practice. Therefore, the PTAB erred by requiring Regents to know that the invention would work prior to showing reduction to practice to prove conception.

The Federal Circuit noted that what does need to be evaluated is how much the final invention differed from the initial conceived idea, despite uncertainty about the successful implementation of the initial idea.  Therefore, the PTAB erred by not considering whether a person of ordinary skill in the art could have reduced the invention to practice based on the Regents’ initial conception of the invention.  The Federal Circuit also discussed the Hitzeman v. Rutter, 243 F.3d 1345 (Fed. Cir. 2001) precedent which provides that a bare hope of a result never before achieved is not sufficient to establish conception.  However, the Federal Circuit distinguished Hitzeman from the present case, because Hitzeman claimed their invention by reciting a particular result of an intracellular process where the method of achieving that result was unknown.  This is different from the present claim because the patent application at hand claims a specific known function of CRISPR-Cas9, and not the result of that function.

Ultimately, PTAB erred by failing to consider routine methods or skill, and focusing nearly entirely on Regents’ perceived experimental difficulties and related statements of doubt.  For remand, the Federal Circuit provided guidance that the appropriate analyses should turn on whether Regents’ scientists had formed the idea of the invention in sufficiently final form that only ordinary skill was needed to reduce it to practice.  More than “general hope” that it would work is needed, but less than knowing with certainty that it would work.

The Federal Circuit also found no errors in the PTAB’s evaluation of written description for Regents’ first two provisional applications because the provisional applications failed to disclose specific conditions for the use of CRISPR-Cas9 in eukaryotic cells.  Broad’s claim construction cross-appeal was also dismissed.

This case reflects how the messy process of scientific research rarely falls cleanly within the lines of patent law.  The opinion can be found here.

By Ann Bernert

This article summarizes aspects of the law and does not constitute legal advice. For legal advice with regard to your situation, you should contact an attorney.