The CRISPR patent dispute between the Broad Institute and the University of California group has not ended, but as of October 10, 2026 Broad is ahead. On March 26, 2026 the US Patent Trial and Appeal Board (PTAB) again ruled that Broad’s scientists were first to invent CRISPR-Cas9 editing in eukaryotic cells, which include human cells. I could find no report of an appeal since. This article covers US law and the US proceeding, with a short note on Europe.
The headline can mislead. The ruling decides who invented one specific thing first. It does not hand Broad every CRISPR patent, and it does not stop companies from developing CRISPR medicines, which many already do under licenses.
The ruling in one paragraph
The fight is Interference No. 106,115, a proceeding the Patent Office uses to decide who invented something first when two parties claim the same invention. On remand from the Federal Circuit, the PTAB found that the California, Vienna and Charpentier group (CVC) had not shown it conceived of a working eukaryotic embodiment before Broad’s reduction to practice, dated by the Board to October 5, 2012 (Berkeley News). It also rejected CVC’s argument that Broad derived the idea from CVC, reasoning that CVC first had to prove its own earlier conception (IPWatchdog).
A timeline of how it got here
| Date | Event |
|---|---|
| December 2012 | Broad files its priority application, aimed at eukaryotic cells |
| March 2013 | CVC files a broader application |
| April 2014 | Broad’s first patent issues |
| February 15, 2017 | PTAB finds no interference in fact in the first proceeding; the Federal Circuit affirms in September 2018 |
| June 2019 | PTAB declares the second interference, covering 14 CVC applications, 13 Broad patents and one Broad application |
| February 2022 | PTAB awards priority to Broad; CVC appeals |
| May 12, 2025 | Federal Circuit vacates the conception finding and remands |
| March 26, 2026 | PTAB again awards priority to Broad |
The earlier dates come from DrugPatentWatch and the Wilson Sonsini client alert.
What the Federal Circuit changed in 2025
The 2025 opinion, written by Judge Reyna for a panel with Judges Hughes and Cunningham, is the legal backbone of everything since (opinion). The court held that the Board had confused conception with reduction to practice: an inventor does not need to know an idea will work for conception to be complete. It also said the Board leaned too heavily on scientists’ expressions of doubt and should have asked whether a skilled person could have reduced the idea to practice with routine methods.
The court did not give CVC everything. It affirmed the Board’s finding that CVC’s first 2012 provisional application lacked adequate written description for the claimed invention, which blocked CVC from using that filing to claim an earlier date by constructive reduction to practice. Broad’s cross-appeal on the meaning of “guide RNA” was dismissed as moot.
In plain terms, the court told the Board to apply a more forgiving standard to CVC. On remand the Board applied it and still found for Broad.
What the 2026 decision does and does not decide
- It decides priority on the interference count, which is CRISPR-Cas9 in eukaryotic cells. Per UC, the result prevents 14 CVC applications from moving to allowance.
- It leaves alone more than 60 US patents and more than 40 non-US patents that UC says CVC holds, covering Cas9 use in all cell types (Berkeley News).
- It does not end parallel interferences involving ToolGen and Sigma-Aldrich, which Wilson Sonsini says remain suspended pending this one. The Board acknowledged those labs also showed eukaryotic Cas9 editing in the relevant period.
That last point is worth stressing. Even a final Broad win would not make Broad the sole owner of CRISPR in practice. The technology rests on a stack of overlapping patents, and any commercial developer has to work out which ones matter for its own product.
Is there an appeal?
Wilson Sonsini outlined CVC’s options after the decision: a request for rehearing within one month, or an appeal to the Federal Circuit within nine weeks. The firm called a successful appeal challenging, because the Board addressed the legal standards on remand and the court usually defers to the Board on facts. By those windows, the deadline for an appeal would have fallen around late May 2026, though that is my reading of the stated periods and not a date any source gave.
UC said it was disappointed by the ruling. I found no report that a notice of appeal or a rehearing request was filed, and I could not confirm either way. Anyone relying on this point should check the docket directly.
Who licenses what
Because ownership is unresolved, licensing has done the real work of letting products reach patients. The map looks like this.
| Patent family | Typical licensing route |
|---|---|
| Broad, MIT, Harvard | Non-exclusive licenses for commercial research; a separate exclusive arrangement for human therapeutics with Editas Medicine, according to DrugPatentWatch (Broad’s own page describes the non-exclusive research model) |
| University of California | Exclusive, sublicensable license to Caribou Biosciences, which has sublicensed widely; Intellia holds the exclusive license for most human therapeutics |
| Charpentier | Licensed separately to CRISPR Therapeutics and ERS Genomics, which also sublicense |
The UC and Charpentier information comes from Berkeley’s 2025 write-up. Hedging between camps is common. The sharpest example is Casgevy, the sickle cell therapy from Vertex and CRISPR Therapeutics that the FDA approved on December 8, 2023. Four days later Vertex took a non-exclusive license to Editas’s Cas9 technology for ex vivo editing of the BCL11A gene in sickle cell disease and beta thalassemia: $50 million upfront, up to $50 million more contingent, and annual fees of $10 million to $40 million through 2034, according to Editas’s 8-K filing.
A company whose own patent family may or may not be enough can pay for insurance on another side. That is a practical answer to legal uncertainty, and it costs real money.
Why the conception ruling reaches past CRISPR
Most patent fights are about whether an idea is new. This one turned on a narrower question: when does an inventor have the idea fully formed in mind? Patent lawyers call that moment conception, and the Federal Circuit’s 2025 holding is a statement about all inventions, not only gene editors.
The court’s position is that doubt about whether something will work does not by itself mean the idea is incomplete. What matters is whether the doubts led to substantive changes in the system. That may sound abstract, yet it matters for any field where early experiments are uncertain, which includes most of biotechnology.
It also explains why the Board’s second look was not a formality. The first decision had been built on the scientists’ own expressions of uncertainty. After the remand the Board had to look at what a practicing researcher with routine techniques could have done, and it concluded that CVC’s 2012 work reflected ongoing experimentation and not a working eukaryotic system, according to the Wilson Sonsini summary.
What about Europe?
The US proceeding does not bind European patent offices. DrugPatentWatch reports that the European Patent Office revoked Broad’s patent EP2771468 on a priority defect, with the revocation upheld on appeal in 2020. I could not verify this against EPO records, so treat it as a lead to check. The wider point stands: the same inventors can win in one jurisdiction and lose in another because each office applies its own rules.
What to watch next
- Whether any appeal or rehearing request surfaces on the Federal Circuit or PTAB docket.
- Whether the ToolGen and Sigma-Aldrich interferences restart.
- Whether patent term restoration for time spent in interference gives the winner extra years of term, a point Wilson Sonsini raised.
- Whether commercial CRISPR therapy developers change their licensing mix after the ruling.
Common questions
Who owns CRISPR?
No single party. Broad leads on eukaryotic priority, CVC holds many patents covering Cas9 in all cell types, and other groups have claims pending. Products are usually launched under a bundle of licenses.
Does the ruling affect existing medicines like Casgevy?
The sources I found describe licensing deals and not any injunction or damages from the ruling. I found nothing showing a marketed product has been stopped.
Why does it matter who invented it first?
In this old-style interference, priority decides who gets the patent claims. The winner can license the technology and collect royalties.
Could the result flip again?
Yes, in principle, if CVC appealed and succeeded. The Federal Circuit has already shown it will correct the Board’s legal standards once.
This is general information about US law, not legal advice.
Sources
- Federal Circuit, Regents of the University of California v. Broad Institute, Nos. 2022-1594, 2022-1653 (May 12, 2025)
- UC Berkeley News, PTAB sides with Broad Institute (March 26, 2026)
- IPWatchdog, PTAB Upholds Priority Win for Broad Institute
- Wilson Sonsini, PTAB Again Rules in Favor of Broad
- UC Berkeley News, Federal appeals court sends CRISPR-Cas9 patent case back (May 12, 2025)
- Broad Institute, Statements and background on CRISPR patent process
- Editas Medicine, Form 8-K (December 12, 2023)
- Vertex, FDA approval of CASGEVY (December 8, 2023)
- DrugPatentWatch, Platform Patents Were Supposed to Protect Gene Editing
