ith the Broad Institute’s big win on Wednesday in its battle over key patents on the CRISPR-Cas9 genome editing technology, everything is now crystal clear. Kidding!
The University of California seemed to be the loser, since the patent judges denied its effort to effectively block the Broad’s patents. But in a call with reporters, Paul Alivisatos, UC Berkeley’s vice chancellor for research, was upbeat that the ruling would allow UC’s patent claims to finally “move forward.” Berkeley biochemist Jennifer Doudna, whose pioneering CRISPR discoveries UC has been trying to patent, pronounced herself “delighted.” Alivisatos dodged a question about why, if UC found the decision so great, it had issued a statement saying it was “considering all of its options,” including an appeal.
That was right before the discussion turned to green tennis balls. (More on that below.) What inquiring minds want to know:
How can UC be happy?
When it filed for patents based on a 2012 discovery by Doudna and her chief collaborator, Emmanuelle Charpentier, it reached for the moon, asking for patents covering CRISPR genome editing “in any setting, including eukaryotic cells and other cell types.” Eukaryotic cells are those with a nucleus, like animal and plant cells — including, of course, human cells. But the 2012 experiments showed only that CRISPR can cut DNA that’s floating in a test tube; they didn’t demonstrate such editing in cells. “It is undisputed that [Doudna’s 2012 paper] does not report the results of experiments using the CRISPR-Cas9 system in a eukaryotic cell,” the patent judges wrote in their ruling.
Nevertheless, UC is putting a brave face on its prospects. “The patent the Broad received is for the use of CRISPR genome-editing technology in eukaryotic cells,” Doudna told reporters after the patent decision. “Ours is for its use in all cells.” (By “ours,” she means the patents that UC filed for in 2012 but which had been held up by the legal proceeding.) The patent judges’ opinion offered UC reason to hope: An “earlier disclosure of a genus does not necessarily prevent patenting a species member of the genus,” it wrote, suggesting that a larger category of something can be patented separately from a subset of it.
What will that mean for licensees of CRISPR patents?
UC believes that any company that wants to use CRISPR to develop human therapies — we’re looking at you, Editas Medicine — will need to license not only the Broad’s patents on eukaryotic cells but also those UC expects to receive on all kinds of cells. “It looks to me as if someone wanting to use the Broad patent would also have to license the UC patent,” agreed law professor Hank Greely of Stanford University. “The UC patent (if granted) would be on any use; the Broad would be on use in eukaryotes. I think someone who wanted to do this in eukaryotes would need to have licenses to both.”
That’s where green tennis balls come in. “It is possible in patent law to get a patent on all tennis balls and for someone else to get a patent on green tennis balls,” said UC attorney Lynn Pasahow. For “all tennis balls,” read “all cells.” For “green tennis balls,” read “eukaryotic cells.”
What sank UC’s chance to invalidate the Broad’s patents?
The judges’ decision that there was no “interference in fact” meant that the CRISPR patents issued to the Broad covered different inventions than those in the patents UC applied for. And crucially, it was not “obvious” how to go from Doudna’s work (CRISPRing free-floating DNA) to Zhang’s (CRISPRing human cells). If the judges found that Zhang’s achievement was “obvious,” given what Doudna did, then it might not be patentable. UC tried with all its might to argue that extending Doudna’s work in this way was obvious.
But the judges extensively cited Doudna’s own statements, mostly to reporters, in concluding otherwise. “We weren’t sure if CRISPR/Cas9 would work in eukaryotes — plant and animal cells,” she said. Her team had experienced “many frustrations” getting CRISPR to work in human cells. Genetic “techniques for making these modifications in animals and humans have been a huge bottleneck in both research and the development of human therapeutics,” she cautioned. The patent judges concluded that “the inventors themselves were uncertain” about making CRISPR work in human cells.
That could send a terrible message to scientists, said Dr. Robert Cook-Deegan of Arizona State University, an expert on legal and ethical issues surrounding biotechnology. “I hope this does not become the poster child that patent offices use to coach scientists not to be honest about uncertainties about their discoveries,” he said. “The fact that Doudna’s quotes were used by the judges mainly tells me Doudna was being honest. I hope scientists will continue to be honest and not succumb to being told they can’t say things that might undermine a broad patenting strategy.”
Please, is it almost over?
Legal experts told STAT that they expect UC to appeal to a circuit court, and Alivisatos kept the door to that wide open. But with the legal meter ticking — it passed $15 million for both sides last summer — patent experts offered a glimmer of hope for an end game. “This is the scenario that should most strongly drive the parties toward some kind of bargain,” said law professor Jorge Contreras of the University of Utah. Once Berkeley gets its patent or patents, “we’ll see whether their patents are required for commercial implementation of CRISPR. If so, companies could need something from both Broad and UC.” That recognition might pave the way to a settlement.
If so, it could and should have happened sooner, experts said. The Broad and UC “should have done a cross-licensing deal,” said Cook-Deegan. “Everyone could have kept more money in their pockets with much [less] scientific rivalry and animus, and they could still have divvied up the rewards.” The result “really is too bad,” he added, “because UC could use the money a whole lot more than the Broad.”
Might this be less momentous than it seems?
Yes. CRISPR-Cas9 is unlikely to be the last genome-editing technology ever discovered. In 2015, Zhang and his colleagues discovered a version called Cpf1, which they’ve now patented and licensed to Editas. “I continue to think the possibility of inventing around the [CRISPR] patents seems very likely,” said Stanford’s Greely. Bacteria “have certainly come up with other ways to reach the same end [of genome editing], ways that aren’t covered by UC’s or the Broad’s claims. That could make either of these patents ultimately of little importance … especially if the licensing conditions give people a strong incentive to come up with invent-arounds.” Science will march on.