CRISPR Patent War: UC Berkeley vs. Broad Institute and the Battle Over Gene Editing’s Crown

In June 2012, a paper appeared in the journal Science that would change biology forever. Its authors—Jennifer Doudna of the University of California, Berkeley and Emmanuelle Charpentier, then at the University of Vienna—described a programmable molecular machine capable of cutting DNA at any specified location with unprecedented precision. The system, adapted from bacterial immune defenses, was called CRISPR-Cas9. With it, editing the genetic code of any living organism became, for the first time in history, practical, affordable, and accessible.

Doudna and Charpentier would receive the Nobel Prize in Chemistry in 2020 for this discovery. But before the Nobel, before the celebration, before the biotech revolution that CRISPR would unleash, came a legal battle of extraordinary complexity and consequence. The question—deceptively simple on the surface—was this: Who owns CRISPR?

The answer, it turned out, would take a decade of litigation to partially resolve, spawn one of the most technically complex patent interference proceedings in USPTO history, and ultimately distribute CRISPR’s IP landscape in ways that continue to shape the $10 billion gene-editing industry today.

The Science: Why CRISPR Was Worth Fighting Over

To understand the patent war, we must first understand why CRISPR-Cas9 was so revolutionary—and why its commercial value was instantly obvious to anyone paying attention.

Before CRISPR, gene editing tools existed but were expensive, slow, and technically demanding. Zinc finger nucleases (ZFNs) and transcription activator-like effector nucleases (TALENs) could both cut DNA at specific sites, but engineering a new ZFN or TALEN to target a new genomic location required months of work and hundreds of thousands of dollars in specialized expertise. CRISPR-Cas9 reduced this to a matter of days and a few hundred dollars in reagents. Instead of building a new protein for each target, CRISPR uses a “guide RNA”—a short RNA molecule that can be cheaply synthesized and programmed to match any DNA sequence. The guide RNA directs the Cas9 protein to the target site, where it cuts both strands of the DNA helix.

The implications were staggering. Correcting genetic diseases. Engineering crops with new traits. Creating animal models of human diseases for drug research. Modifying immune cells to attack cancer. Building entirely new organisms with designed genomes. Every application of gene editing—therapeutic, agricultural, industrial—was suddenly transformed by CRISPR’s simplicity and versatility.

This was worth billions. Probably tens of billions. And two scientific teams were about to collide over who had gotten there first.

The Contestants: Berkeley and the Broad

The University of California, Berkeley team, led by Jennifer Doudna, filed its first CRISPR-Cas9 patent application in May 2012, based on work demonstrating that CRISPR-Cas9 could cut DNA in vitro—in purified, cell-free biochemical systems. The application claimed the use of CRISPR-Cas9 in “any environment”—a broad formulation that the Berkeley team argued encompassed both cell-free systems and the more complex environment of living eukaryotic cells (animal and plant cells).

The Broad Institute of MIT and Harvard team, led by Feng Zhang, filed its first CRISPR-Cas9 patent application in December 2012—seven months after Berkeley—but paid for expedited examination, receiving the first issued CRISPR-Cas9 patent (US 8,697,359) in April 2014. The Broad’s application focused specifically on the use of CRISPR-Cas9 in eukaryotic cells—the cells of animals and plants, which are far more commercially important for therapeutic and agricultural applications than bacterial cells.

The stage was set for a fundamental question of patent law: Does a pioneering disclosure of a technology’s basic operation in simplified conditions automatically claim all applications of that technology, including applications in more complex conditions that the pioneer may not have specifically reduced to practice?

The Interference Proceeding: Patent Law’s Nuclear Option

Patent interference proceedings are among the most complex and expensive procedures in the US patent system. Under the pre-America Invents Act (pre-AIA) patent system—which governed patents filed before March 16, 2013—the US used a “first to invent” system rather than the “first to file” system used by most of the world. When two parties both claimed the same invention, the USPTO’s Patent Trial and Appeal Board (PTAB) conducted an “interference” to determine who was the true first inventor.

Both Berkeley and the Broad had filed their key applications before March 16, 2013, putting them in the pre-AIA “first to invent” world. Berkeley filed first but had not yet received a patent. The Broad received its patent first. Berkeley initiated an interference proceeding to determine who had first invented the use of CRISPR-Cas9 in eukaryotic cells.

The critical legal question was whether there was an “interference in fact”—whether Berkeley’s claims and the Broad’s claims were directed to the same invention. If Berkeley’s claims (directed to CRISPR-Cas9 “in any environment”) inherently encompassed eukaryotic cells, there was an interference in fact, and the PTAB would need to determine who had first reduced this invention to practice. If Berkeley’s claims were limited to cell-free environments and did not obviously extend to eukaryotic cells—if getting CRISPR to work in the more complex environment of animal cells required inventive skill that Berkeley had not demonstrated—then there was no interference in fact and the Broad’s patents on the eukaryotic application were separate inventions.

The January 2017 Decision: Broad Wins Round One

In January 2017, the PTAB ruled in a decision that sent shockwaves through the biotechnology community: there was no interference in fact. The Board found that going from cell-free CRISPR-Cas9 to CRISPR-Cas9 operating in eukaryotic cells was not an obvious extension—it required significant additional inventive work, work that Berkeley had not publicly disclosed in its 2012 paper or patent application. The Broad’s eukaryotic-cell patents were therefore a separate patentable invention from Berkeley’s cell-free work.

Translated into plain English: Berkeley invented the basic CRISPR-Cas9 system. The Broad invented its specific use in the cells that matter most commercially. Both inventions were patentable, and they were different enough that the Broad could hold patents on the commercially critical application even though Berkeley had discovered the underlying mechanism first.

Berkeley appealed. In September 2018, the Federal Circuit affirmed the PTAB’s decision in a terse opinion that emphasized the deference owed to the Board’s factual findings about what was or was not obvious to a skilled biologist in 2012.

The Second Interference: An Even Deeper Dive

The first interference resolved one question but left others open. Berkeley had continued prosecuting its patent applications, and some of those applications resulted in issued patents that more specifically claimed eukaryotic cell applications. The Broad challenged these patents through inter partes review (IPR) proceedings—the post-AIA procedure that replaced interference for patents issued after 2013.

A second interference proceeding (Interference No. 106,127), focusing on different sets of claims, was initiated in 2019. This proceeding produced yet another round of extraordinarily detailed technical and legal argument about the priority of invention—who had first demonstrated that CRISPR-Cas9 would work in eukaryotic cells, and what laboratory records, email communications, and research notebooks could prove it.

In March 2022, the PTAB issued its decision in the second interference: the Broad Institute won again, on similar grounds. The Board found that Feng Zhang and his colleagues at the Broad had first demonstrated CRISPR-Cas9 functioning in a eukaryotic cell, and that Berkeley’s team had not shown prior reduction to practice for this specific application.

The Federal Circuit once again affirmed, in a February 2022 decision, cementing the Broad’s position on the core eukaryotic CRISPR patents.

But Wait—Berkeley Also Wins

The narrative of “Broad wins the CRISPR patent war” is accurate but incomplete. The US patent landscape for CRISPR-Cas9 is not a winner-take-all system, and Berkeley’s patent portfolio—while it lost the eukaryotic application battle—remained valuable and extensive.

Berkeley holds issued US patents covering the fundamental CRISPR-Cas9 machinery: the complex of Cas9 protein with guide RNA, the mechanism of target DNA recognition, and the basic molecular biology of the system. These foundational patents cover technologies that the Broad’s eukaryotic-application patents depend upon: you cannot practice CRISPR in eukaryotic cells without using the fundamental mechanism that Berkeley’s patents describe.

This means that in practice, anyone seeking a complete license to practice CRISPR-Cas9 therapeutics needs licenses from both Berkeley (for the foundational machinery) and the Broad (for the eukaryotic cell application). The practical result is a patent landscape that neither party controls exclusively—a situation that has led to extensive cross-licensing negotiations and, for third-party companies, sometimes complex freedom-to-operate analyses involving both patent estates.

The Global Picture: Europe and Asia Take Different Paths

The CRISPR patent story in Europe played out quite differently. The European Patent Office (EPO) examined CRISPR patent applications under a different legal framework—particularly the EPO’s strict enablement and written description requirements, which assess whether the patent application’s disclosure sufficiently supports the breadth of the claims.

In Europe, Berkeley’s CRISPR patents have generally fared better. The EPO granted key Berkeley CRISPR patents covering a broad scope of CRISPR-Cas9 applications, including eukaryotic cell uses. The EPO’s determination of patentability focuses on what was disclosed in the application itself, with less emphasis on the “who invented first” question that dominated the US interference proceedings.

The Broad’s European patents were challenged through EPO opposition proceedings—the European equivalent of IPR—by a coalition of opponents including Berkeley and several biotech companies. The outcomes have been mixed, with some Broad patents maintained with narrowed claims and others revoked. The European CRISPR patent landscape as of 2026 remains complex and actively litigated.

In Asia, the picture varies by jurisdiction. Chinese courts have had to apply Chinese patent law to CRISPR disputes, and given China’s enormous investment in CRISPR research and the many Chinese biotech companies developing CRISPR therapeutics, the Chinese patent landscape for CRISPR is commercially critical. Chinese patent filings by both Berkeley and Broad affiliates, as well as by Chinese institutions like Zhejiang University, have created a complex web of rights that Chinese CRISPR companies must navigate.

Japan’s patent system, with its own examination standards and prosecution history, has also produced a distinct CRISPR patent landscape. The Japanese Patent Office (JPO) has issued CRISPR patents to both Berkeley-affiliated and Broad-affiliated applicants, though with claim scopes that reflect the JPO’s specific requirements for enablement and support.

Commercial Consequences: The Race to License CRISPR

The patent battle has had profound commercial implications, shaping the structure of the entire gene-editing industry.

Berkeley’s licenses to the foundational CRISPR technology are managed through the Innovative Genomics Institute (IGI) and through UC Berkeley’s licensing office. Berkeley has pursued a licensing strategy that distinguishes between therapeutic (human health) applications and agricultural applications, with different terms and different licensees for each.

For therapeutic applications, Berkeley’s foundational CRISPR patents have been licensed non-exclusively to multiple companies—allowing competition in the CRISPR therapeutics space while generating royalty revenue for UC Berkeley. Editas Medicine (co-founded by CRISPR pioneer Feng Zhang himself), CRISPR Therapeutics, Intellia Therapeutics, and dozens of smaller companies all required licenses to practice CRISPR in their therapeutic programs.

The Broad, through a licensing arrangement with Editas Medicine, gave Editas exclusive rights to certain Broad CRISPR patents in certain therapeutic areas. This exclusive license became highly controversial when other CRISPR companies (particularly CRISPR Therapeutics, which was co-founded by Emmanuelle Charpentier and licensed Berkeley’s technology) found themselves potentially blocked from certain applications by Editas’s exclusive rights.

The result was a complex commercial ecosystem where different CRISPR companies had different patent positions, required different license combinations, and had different freedom-to-operate in different therapeutic areas. For investors and business developers trying to assess CRISPR companies, understanding the patent landscape became a fundamental due diligence requirement.

The Nobel Prize and Its Legal Shadow

In October 2020, the Nobel Committee awarded the Chemistry Prize jointly to Jennifer Doudna and Emmanuelle Charpentier for “the development of a method for genome editing.” The committee’s citation emphasized the elegance and simplicity of CRISPR-Cas9—properties that the Nobel citation attributed to Doudna and Charpentier’s foundational work.

Feng Zhang was conspicuously absent from the Nobel. His advocates argued that Zhang’s work in demonstrating CRISPR in eukaryotic cells was equally deserving of recognition—that the foundational biochemistry and the therapeutic application are both essential contributions to the CRISPR revolution. Zhang’s supporters pointed to the US patent proceedings’ findings that eukaryotic CRISPR required significant additional inventive work as evidence that Zhang’s contribution was independently praiseworthy.

The Nobel decision did not resolve the patent dispute—it is an award for scientific achievement, not a patent determination—but it amplified the narrative that Berkeley’s foundational work was the essential discovery, lending cultural weight to Berkeley’s legal position even as the courts had found otherwise on the specific eukaryotic application question.

CRISPR Therapeutics: From Patent to Medicine

Whatever one thinks of the patent dispute, the downstream therapeutic applications of CRISPR are transforming medicine. The first CRISPR-based human therapies reached regulatory approval in late 2023 and 2024, beginning with treatments for sickle cell disease and beta-thalassemia.

Casgevy (exagamglogene autotemcel), developed by Vertex Pharmaceuticals and CRISPR Therapeutics, received FDA approval in December 2023—the first CRISPR-based therapy approved anywhere in the world. Lyfgenia (betibeglogene sparticicvec), a gene therapy from bluebird bio using a different mechanism, received concurrent approval. For patients with these devastating blood disorders, these therapies offered the potential for functional cures.

Casgevy’s approval was a commercial and scientific milestone—but it also triggered licensing royalty streams that flowed along the patent landscape established by the Berkeley/Broad litigation. Understanding who holds what CRISPR patents, and who licensed from whom, is essential to understanding the economics of CRISPR therapeutics companies.

As more CRISPR therapies move through clinical trials—for cancer, genetic eye diseases, HIV, and dozens of other conditions—the commercial stakes of the foundational patent dispute will only grow. The cumulative royalty burden on commercial CRISPR products, if it is too heavy, could impede the very therapeutic revolution that the underlying science made possible.

Alternative CRISPR Systems: The Race Beyond Cas9

The CRISPR-Cas9 patent dispute has also accelerated research into alternative CRISPR systems—Cas12, Cas13, CasX, prime editing, and base editing technologies that use different proteins and mechanisms to achieve gene editing with different properties. Some of these alternative systems were developed specifically to avoid the patent thicket around Cas9; others offer genuinely superior properties for specific applications.

Prime editing, developed by David Liu at the Broad Institute, allows precise insertion, deletion, and substitution of DNA sequences without causing double-strand breaks—potentially safer and more precise than conventional CRISPR-Cas9. Base editing, also from Liu’s laboratory, makes single-letter changes to the DNA sequence without cutting. Both technologies have their own patent estates, owned largely by the Broad, with commercial licenses going to Beam Therapeutics and Prime Medicine respectively.

The diversification of CRISPR technology means that the Berkeley/Broad dispute, while central to the first generation of CRISPR therapeutics, will not determine the entire gene-editing IP landscape. Future gene-editing therapies may rely on technologies with entirely different patent landscapes—which may ultimately be more friendly to commercial development than the contested Cas9 patent territory.

Japanese CRISPR Research and the IP Challenge

Japan has been a significant contributor to CRISPR research, with groups at institutions including the University of Tokyo, Osaka University, and the RIKEN institute publishing important work on CRISPR mechanisms and applications. Japanese biotech companies including Takeda Pharmaceutical, Astellas, and several startups have pursued CRISPR-based therapeutic programs.

Japanese CRISPR researchers and companies must navigate the global patent landscape that Berkeley/Broad litigation has created—obtaining licenses from both patent estates for any commercial therapeutic program. The Japan Patent Office’s examination of CRISPR applications has produced Japanese patents for various stakeholders, and the commercial licensing structure in Japan may differ from the US and European structures based on the specific Japanese patent rights available.

Japan’s agricultural sector is also affected by CRISPR IP developments. Japan has approved several genome-edited agricultural products under a regulatory framework that distinguishes between gene-edited organisms (which may not require the same approvals as GMOs) and traditionally modified organisms. As CRISPR-edited agricultural products reach the market, the IP rights governing them—including patents held by both the Berkeley and Broad estates, as well as by agricultural biotech companies—will determine licensing costs and competitive dynamics.

Lessons for the Patent System: Did the Law Get CRISPR Right?

As an IP detective, I find myself asking whether the patent system handled the CRISPR dispute well. The answer is complicated.

On the positive side, the patent system provided a mechanism for resolving a genuine priority dispute through detailed factual analysis of scientific records, notebooks, and disclosures. The PTAB interference proceedings, however technical and expensive, reached reasoned conclusions about what each team had actually invented and demonstrated. This is better than pure commercial chaos.

On the negative side, the complexity and duration of CRISPR patent proceedings—ten-plus years, two major interference proceedings, multiple IPRs, federal circuit appeals—imposed enormous transaction costs on the entire gene-editing industry. Every company developing CRISPR therapeutics had to budget for legal uncertainty, maintain watch on the proceedings, and potentially delay commercial decisions pending resolution.

The fundamental problem is that the US patent system’s rules for determining inventorship in pioneer inventions—inventions that create new technological paradigms rather than incremental improvements—are imperfectly designed. When two research groups independently make related discoveries at roughly the same time, the question of who “owns” the resulting commercial opportunity is ultimately a policy question as much as a factual one. The interference proceeding provides one framework for answering it; other countries use different frameworks, leading to different outcomes in different jurisdictions.

Whether any framework would have produced clearly “better” results for both science and society is genuinely uncertain. What is clear is that the CRISPR patent dispute will remain a central case study in patent law education for decades—and that understanding it is essential for anyone working at the intersection of biotechnology and intellectual property.

Conclusion: The Molecule That Changed IP Law

CRISPR-Cas9 is one of the most significant scientific discoveries of the 21st century. It has generated, in a relatively short time, a patent dispute that touches almost every fundamental question of patent law: What is an invention? Who is an inventor? What does it mean to reduce an invention to practice? How broadly does a pioneering disclosure claim? What policy goals should patent law serve in fast-moving scientific fields?

Jennifer Doudna and Feng Zhang both made real contributions to one of science’s greatest tools. The legal system has tried, imperfectly but earnestly, to draw lines between those contributions. What emerges is a reminder that the patent system is a human institution operating in the face of scientific complexity and commercial stakes that frequently exceed its capacity to provide clean, satisfying resolutions.

The gene-editing revolution continues. The patents will be litigated for years to come. And the patients who will eventually be cured by CRISPR therapies may not know—or care—who holds the patent on the molecular scissors that saved their lives.

But we IP detectives care. And we’ll keep watching.

探偵くん returns to the lab bench—with a warrant.