Could the Whole CRISPR Patent Kerfuffle Have Been Completely Avoided?

By Kristen V Brown on at

For the better part of the last three years, the introduction of the most powerful gene editing technology ever invented has been marred by a nasty patent battle. The two groups of scientists involved, each contributing significantly to the future of genetic engineering, are pitted against each other in a bitter contest for glory and fortune.

In one corner: UC Berkeley, which argued that seminal 2012 work revealing that a bacterial immune system called CRISPR could be repurposed to edit genes ultimately paved the way for CRISPR to be used to genetically alter animal DNA. In the other corner: The Broad Institute of Harvard and MIT, which six months later actually adapted CRISPR to engineer plant, animal and human cells. At issue: Who deserves credit for the original recipe that led to using CRISPR-Cas9 to edit mammalian DNA.

In February, a US Patent Office appeal board raised the Broad Institute’s arm in victory. Since then, this US case has been stuck in appeals within the Federal Circuit (in Europe, the question of who will ultimately win this patent battle is also up for debate). But if both sides had adhered to a controversial 1997 US patent rule known as the written description doctrine, the whole mess probably could have been avoided, argue two researchers in a new Science opinion.

“A patent is an economic instrument, not an instrument for glory,” Arti Rai, co-director of the Duke Law Center for Innovation Policy and an author on the paper, told Gizmodo.

The issue, she said, is that Berkeley (and perhaps also Broad) made claims in their patents that were simply too broad. As written, the patent code already requires that if you patent something, you have to include “full, clear, concise, and exact terms” that would allow anyone with the prerequisite skills to replicate whatever it is that you did. Twenty years ago, the Federal Circuit court made it clear that original patent claims can be invalidated if the patent application does not disclose how the operation of complex new technologies would actually work. In theory, this means you can’t patent the genetic engineering of mammalian cells with CRISPR if you haven’t actually tried to edit mammalian cells with CRISPR yourself.

“You’re supposed to describe in your patent what you have already done,” said Robert Cook-Deegan, Rai’s co-author and a researcher at Arizona State University’s School for the Future of Innovation in Society. “That means that Berkeley would have had to file patent applications specifically describing editing eukaryotes.”

Part of the problem is that the written description doctrine is controversial and inconsistently enforced.

Ariad Pharmaceuticals Inc.’s 2010 patent infringement suit against Eli Lilly, which is a crucial affirmation of the written description doctrine, is a good example. Harvard and MIT scientists had identified a cell-signalling pathway related to the body’s immune responses to an infection known as NF-kB. The patent the scientists licensed to Ariad, though, claimed a method for modulating that pathway, though they had yet to identify the molecules that modulated it. A jury sided with Ariad, claiming that Eli Lilly had manufactured two drugs that infringed upon it, but a Federal Circuit court overturned the finding, invoking the written description doctrine.

In a dissenting opinion, though, a judge articulated a criticism held by many—that a written description doctrine separate from what’s already included in the patent code is just unnecessary. The sentiment was that in expanding the requirements of a written description, the court had overshot its goal and created something that created too many hoops for innovators to jump through that could be used to invalidate almost any patent claim. The crux of the criticism was that all this might hamper innovation and prevent innovators from reaping the benefits they ultimately deserved. (The original written description portion of the patent code was used only to police claims added after an original filing date, not to test whether an inventor actually possessed the ability to do everything described in their original filing.)

Those arguments, though, persist today and the inconsistent application of the doctrine has made it such that institutions like Berkeley have a decent shot at making a successful claim to broad discoveries.

“The courts should uphold the written doctrine consistently so people stop trying to get around it,” said Cook-Deegan. “Then we can quit paying all these lawyers to dispute patents.”

“The weight of evidence suggests that broad knowledge monopolies even in the hands of universities can hinder scientific progress.”

Even the Broad patents, the paper points out, may cover too much inventive territory because it covers all Cas9 proteins, even though there are many such proteins and the work did not cover them all.

It makes sense. After all, broad patents mean more opportunities to reap financial reward.

But overly broad patents, the paper argues, are inefficient and wasteful in that they inevitably lead to long, drawn out patent battles like the one over CRISPR.

“The bigger the prize at the end of the race, the more likely we will see inefficiency,” they write.

If awarded, broad patents also concentrate economic wealth and power over essential scientific knowledge.

“The weight of evidence suggests that broad knowledge monopolies even in the hands of universities can hinder scientific progress,” the paper argues.

The problem of broad licenses is exacerbated by the Bayh-Dole Act of 1980, which was intended to turn patents into a way of commercialising federally funded research and allow research institutions to profit big off their inventions. This has led to complicated licensing agreements that assign ownership and access of broad technological inventions to a few. Both Berkeley and Broad, for example, have doled out exclusive licenses to companies affiliated with the scientists behind the work and anyone who hopes to commercialise those inventions must pay to sublicense the technology from them.

“We would not begrudge them scientific glory. But that’s different from claiming a patent.”

The bigger issue, though, may be the cultural significance associated with patents. Patents are a trophy, an invisible symbol that convey credit for an innovation. Legal rights get confused with scientific credit. But conflating credit and economic ownership, said Rai, is a bad idea.

“For Berkley to make the claims that it did was a great conceptual leap,” said Rai. “We would not begrudge them scientific glory. But that’s different from claiming a patent.”

If it upholds the patent board decision, the Federal Circuit court may wind up invoking the written description doctrine. In the case, a key legal question is whether at the time Berkeley filed its patent, eukaryotic applications were not just “obvious to try” but also had a “reasonable expectation of success.” The patent board held that Berkeley’s breakthrough had not made it scientifically “obvious” that Broad’s work would succeed. That meant there was no overlap in what Berkeley and Broad each claimed in their patents.

Of course, some have argued that a technology as powerful as CRISPR should be owned by no one at all.