The question of who owns the foundational CRISPR-Cas9 patents has a precise legal frame: it is a priority dispute, and as of the most recent appellate ruling it is not finally resolved. On one side are the Regents of the University of California, the University of Vienna, and Emmanuelle Charpentier; on the other are the Broad Institute, the Massachusetts Institute of Technology, and Harvard. The Broad Institute holds issued U.S. grants directed to CRISPR-Cas9 systems and methods in eukaryotic cells — for example, U.S. Patent 11,578,312, “Engineering and optimization of systems, methods, enzymes and guide scaffolds of CAS9 orthologs and variants for sequence manipulation” (issued February 14, 2023; inventors including Feng Zhang; CPC C12N 9/22 and C12N 15/11), and the later U.S. Patent 12,168,789 in the same family (issued December 17, 2024). Both abstracts describe “methods of directing CRISPR complex formation in eukaryotic cells.”

The dispute reached the U.S. Court of Appeals for the Federal Circuit, which issued its opinion in The Regents of the University of California v. The Broad Institute, Inc. on May 12, 2025. The case concerns which party first invented a single-guide CRISPR-Cas9 system that edits or cleaves DNA in eukaryotic cells — the commercially decisive question, because eukaryotic application is what enables editing in human and other animal cells. The court did not declare a winner. It affirmed-in-part, vacated-in-part, and remanded, instructing the Patent Trial and Appeal Board to reconsider the issue of conception under the correct legal standard.

"We hold that the Board legally erred by conflating the distinct legal standards for conception and reduction to practice."— U.S. Court of Appeals for the Federal Circuit, May 12, 2025, source

What the Federal Circuit actually decided

The legal core of the 2025 opinion is about conception — the formation in an inventor's mind of a definite and permanent idea of the complete invention. The Board had ruled against UC's group on priority, and UC argued the Board legally erred by requiring its scientists to know their invention would work before they could be credited with conceiving it. The Federal Circuit agreed there was an error of law, holding that the Board conflated the separate standards for conception and for reduction to practice. The court vacated the Board's conception determination and remanded for the Board to reconsider it under the corrected standard. This is a procedural posture, not a holding on who invented first: the priority question goes back to the Office.

That distinction matters for anyone reading the dispute. A vacate-and-remand does not award the contested claims to either side. It identifies a legal defect in how the lower tribunal reached its conclusion and requires a fresh analysis. Until the Board completes that reconsideration — and any subsequent appeal — the allocation of priority for the single-guide eukaryotic CRISPR-Cas9 claims at the heart of the case remains open. Statements that one institution “won” the foundational patents overstate where the record stands after May 2025.

The conception point deserves a closer look, because it is the hinge of the whole appeal. Conception, in patent law, is the formation in the inventor's mind of a definite and permanent idea of the complete and operative invention; reduction to practice is the later step of actually building or performing it (or filing an enabling application that constructively does so). These are separate inquiries with separate proof standards. The Federal Circuit's holding is that the Board treated them as one — effectively demanding that UC's scientists already know their single-guide system would work in eukaryotic cells before crediting them with having conceived it. Because that is the standard for reduction to practice rather than for conception, the court found legal error and vacated the conception finding. On remand the Board must apply the conception standard on its own terms, which could change the priority analysis.

Why both portfolios can coexist for now

It is also possible for the Broad to hold issued, enforceable eukaryotic CRISPR patents while the priority contest continues, because the interference and appeal address whether certain UC and Broad claims interfere and which party has priority over the overlapping subject matter — not whether every Broad grant is invalid. The Broad's issued grants, such as the 11,578,312 and 12,168,789 patents above, remain on the register as granted rights regardless of the remand. The unresolved question is narrower: priority over the specific contested single-guide eukaryotic claims.

For licensing professionals and litigators, the practical reading is to track the matter by its procedural stage rather than its headlines. The foundational CRISPR estate is split across institutions, the Broad holds issued eukaryotic grants, and the priority question that could reallocate the most contested claims was sent back to the Patent Trial and Appeal Board in May 2025. What happens on remand — and whether it is appealed again — is what will ultimately answer the ownership question the case poses.

The commercial stakes are what keep this dispute in view despite its procedural complexity. Foundational CRISPR-Cas9 patents underpin licensing programs that reach across therapeutics, agriculture, and research-tool markets, and the allocation of priority over the single-guide eukaryotic claims affects which institution's licensees stand on the stronger footing for that subject matter. Companies practicing CRISPR have in many cases taken licenses across both estates precisely because the priority question has remained unresolved — a rational response to legal uncertainty rather than a concession by either side. Until the Patent Trial and Appeal Board completes the conception analysis the Federal Circuit ordered, and any further appeal runs its course, the prudent reading is that the foundational estate is shared and contested, with issued grants on the Broad side and an open priority question that the courts have expressly declined to treat as settled.

This article reports the procedural posture and the cited record; it is not legal advice, and it draws no conclusion about who invented first, a question the courts and the Patent Office have not finally resolved.