Skip to main content
Premium Trial:

Request an Annual Quote

CRISPR License Agreements Prohibit Misuse, but Enforcement is Challenging


NEW YORK (GenomeWeb) – When He Jiankui announced to the world in late November that he had edited the germline genomes of twin embryos in order to make them impervious to HIV — and then implanted those embryos in their mother so they could fully develop and be born — the scientific research community largely reacted with horror.

Researchers decried the seemingly shoddy editing work (in fact, He's work has never been fully explained as his paper has yet to be published), but they primarily slammed his work as "stupendously unethical" and even "eugenic" in nature. In the wake of the revelation, the CRISPR research community called for immediate moratoriums on germline genome editing.

In mid-December, the World Health Organization announced that it planned to establish a global, multi-disciplinary panel of experts to examine the scientific, ethical, social, and legal implications of both somatic and germline genomic editing. The panel — whose members were appointed in February and who will meet for the first time on March 18 — will review the current research and review social attitudes towards genome editing technology, then advise the WHO on appropriate oversight and governance mechanisms.

"Core to this work will be understanding how to promote transparency and trustworthy practices and how to ensure appropriate risk/benefit assessments are performed prior to any decision on authorization," the agency said in a call for comment.

Of course, ensuring appropriate risk/benefit is the bread and butter of commercial enterprise. And these days, one would be hard-pressed to find a company that doesn't talk about transparency and earning the trust of its customers. So, finding out whether He had signed a license agreement for CRISPR IP — a practice that has become increasingly popular in the industry — became almost as interesting to some amateur detectives as finding out whether he'd had help from other researchers or where he'd gotten his funding.

Reports have emerged that scientists such as Nobel laureate Craig Mello, Stanford bioethicist William Hurlbut, gene-editing researcher Matthew Porteus, and biophysicist Stephen Quake may have known about He's work before he publicized it, to varying degrees. Indeed, Rice University's Michael Deem has been implicated in the experiment itself.

And while a preliminary investigation by China's National Health Commission and Shenzhen's office of Health and Family Planning, as well as by the Southern University of Science and Technology where He had worked, found that he performed his work illegally and raised funds for it on his own, other reports suggested that the Ministry of Science and Technology, the Shenzhen Science and Technology Innovation Commission, and the university funded the research, though they may not have known how the money was going to be used.

As for the technology itself, Wired Magazine reported in November that the consent forms He gave to the parents who participated in his experiments showed that he bought the Cas9 nucleases he used from Thermo Fisher Scientific and the synthetic guide RNAs he used from Synthego. His use of Synthego's products would have violated the company's terms of sale, which state that the gRNAs are for research use only, and not for human or therapeutic use.

Further, MIT Technology Review reported in December that the Broad Institute had signed a licensing deal with He in 2016 to market certain CRISPR IP in China. Although the researcher seemingly didn't use the institute's nucleases and other products to conduct his experiments, the He fiasco has raised a question about the ubiquity of CRISPR licensing deals and the technology's ready availability to just about anyone.

For its part, the Broad has pointed to its licensing agreement, which also has a provision prohibiting germline genome editing, reading in part, "The Field shall expressly exclude (a) any human or clinical use including, without limitation, any administration into humans or any diagnostic or prognostic use, (b) any human germline modification, including modifying the DNA of human embryos or human reproductive cells…"

According to the Broad's Chief Business Officer Issi Rozen, these terms have been included in the licensing agreement since the institute started licensing its CRISPR IP to spinout company Editas Medicine in 2013.

"From the beginning, we thought it's premature to go and edit germline DNA or embryos, and we put the restriction on. So, that was, of course, for human use," Rozen said. "Also, when we started doing agricultural uses of CRISPR, we put some public policy statements [in the licensing terms.] For example, we restricted modifications of tobacco for smoking, we restricted gene drives. In each category, we were thinking about the main public policy questions, and we try through the licensing to restrict the ability of our licensees to do that."

But rather than using the prohibition merely to sue licensees who stray from the terms, the contract is perhaps more of a philosophical statement of values, laying out the Broad's place in the public debate surrounding genome editing in general, and its effect on humans in particular.

"It will come down to regulatory authorities, and governments, and institutions to enforce [policies on germline editing]. We are not the regulatory authority, but we do have a voice in the public debate that I think is an important voice… and that restriction is in line with public policy statements and in line with a view of where we are right now with the technology as responsible scientists," Rozen said. "So, it's less because we think that if there is some scientist somewhere in the world that does that, we can go after them. That's probably very unlikely. But it is very visibly telling the world how we view this and what is our position on it."

Technically, he added, if the Broad were to find that someone had breached the terms of the licensing agreement, the institute would terminate the contract. It would also be within its rights to sue for damages, though that wouldn't be as easy as ending the agreement. He also said that the Broad has not yet had occasion to terminate a licensing contract for breaching the prohibition on germline genome editing.

Caribou Biosciences, which has a large CRISPR IP portfolio — including a license to the foundational CRISPR-Cas9 technology from the University of California and the University of Vienna — licenses its technology to other companies. Its various licensing agreements differ in certain language depending on the technology in the agreement, but they mostly have one thing in common: they forbid the editing of human embryos.

In this limited use label license Integrated DNA Technologies signed with the company, for example, IDT agreed to "only use the Product and/or Related Materials for its research use," and to exclude any activity "in a medical application where a result is transmitted to a patient and/or caregiver for clinical decisions" or "for editing the genomes of human embryos," among other things.

"At Caribou, we don't edit human embryos for any purposes. And in several of our licensing agreements, we have explicitly forbidden the licensee, and, if it's relevant, the licensee's customers from using our technology to edit human embryos," Caribou Cofounder and CEO Rachel Haurwitz said. "It's an issue that's clearly front and center, and has been for quite some time. We think it's an important topic to discuss and understand, and for us, there was a very clear answer. But because of our business model, we think it impacts how we engage with the broader community who are using these technologies. So, it was a natural extension for us to include that type of restriction in license agreements."

She also believes that enforcing such contracts against bad actors could mean filing suit for patent infringement. If a customer were to use CRISPR products it licensed from Caribou to edit human embryos, that would mean using products in a way it wasn't licensed to do, and that could mean infringing the intellectual property, she theorized.

"At that point, depending on which patents we're talking about, whoever the owners are would have the right to enforce their patents," Haurwitz said. "Certainly, the University of California, the University of Vienna, and Emmanuel Charpentier, who jointly co-own foundational CRISPR-Cas9 IP, have patents issued in multiple jurisdictions around the world, including in China." That might make enforcement with international customers a bit easier.

For certain companies, however, the #CRISPRbabies controversy (as it came to be known on Twitter) was a wake-up call. Inscripta — the startup company that made a name for itself by releasing a new CRISPR enzyme it had developed to the research community free of charge — has now updated its licensing contract terms to explicitly exclude germline genome editing, but its previous contracts didn't include this language.

"Use of MAD7 requires adherence to the terms and conditions of our license agreement as well as applicable laws and regulations. Since germline editing of human embryos for pregnancy is illegal in most jurisdictions, Inscripta's original license agreement did not explicitly exclude this field of use," Inscripta CEO Kevin Ness said in an email. "In light of Dr. He's announcement, we updated our license agreement to reinforce the exclusion in this field of use."

The company's license terms now specifically exclude "1) Sale or re-sale of MAD7, including as part of a therapeutic product; 2) continued use of MAD7 in a commercial manufacturing process; and 3) use of the product for editing human embryos (collectively “Excluded Fields”).  For clarity, the use of MAD7 to create a commercial product that does not physically contain MAD7 at the time of sale and does not require continued use of MAD7 for manufacture is not considered an Excluded Field of Use."

Ness also noted that the company would revoke the license of any customer who violated the terms of the agreement. "Should such a violation be brought to our attention, Inscripta would immediately open a dialogue with the party and determine appropriate next steps," he added. "We believe these issues will be effectively managed through dialogue; however, should this not result in the proper outcome, we do reserve the right to pursue all legal avenues to remedy the violation of the licensing terms."

The Broad's Rozen said he wouldn't be surprised to see more organizations start to add such prohibitions to their CRISPR licensing contracts, if they don't already have them. "All of a sudden, people are paying attention. The reason why people didn't put it in the contracts is because they just didn't think about it. It's not that they were supportive of [germline genome editing], it just wasn't something that they thought about at the time they granted licenses," he said.

"I think symbolically it's a good statement and it sets expectations. Certainly, the researcher who might use it outside the scope of the license can't be under any illusion that he or she is using it appropriately," concurred Duke University law professor Arti Rai. "The only question is whether they care about sanctions and what sanctions might actually be delivered."

From a legal perspective, the extent of the punishment that could result from breaking such a provision is really the question for Rai. First, she noted, it most likely wouldn't be easy to enforce such a prohibition overseas, especially if a CRISPR firm is based in the US, because it would have to monitor the actions of its customers very closely in order to catch them in the act.

Second, Rai added, "some of these firms might try to play hardball and say, 'This is patent infringement because you're violating the terms of the license.'" But arguing patent issues from a licensing perspective can be tough, especially from one country to the next, she said. And arguing for specific damages would make it even harder.

"So, then you just break the contract, either under a breach of contract or patent infringement theory," she noted. "I'm not sure what your damages would be exactly, because even in the US we don't have statutory damages. And so, you could try to break the contract and enjoin for future use, but if they'd already used [your product], I don't know that you'd be able to go back and undo that use."

In the end, Rai added, it would seem that the primary purpose of the license prohibitions is to send a clear message of what a company expects from its customers, and what boundaries it's setting.

Rozen concurred. "I don't think that contracts will address this issue. Of course, they enforce against irresponsible licensees, but it's more of a societal debate than [one about] regulatory, and governmental, and institutional policies here," he said. "Institutions have to have policies for the behavior of faculty. There has to be a public debate and a regulatory debate — what's safe to do at what point in time. And at some point, we need to look again and decide what our policy is. That would be more effective than the handful of companies granting licenses trying to enforce it directly."