Science

An Outdated Law Will Decide the CRISPR Patent Dispute

On Tuesday, a crowd piled into a patent hearing that will be among the last of its kind.

By Sarah Zhang
The United States Patent and Trademark Office headquarters in Alexandria, VA.  (Alex Brandon / AP)
Saved Stories

ALEXANDRIA, Virginia—Nobody could recall such a long line at the U.S. Patent and Trademark Office. On Tuesday, more than an hour before the scheduled hearing for a bitterly contested patent dispute over the gene-editing tool CRISPR, a line of lawyers, journalists, and biotech industry execs had snaked through the lobby and across the giant Christmas tree that bedecks government-agency headquarters at this time of the year.

Of course, the dispute over CRISPR is no ordinary patent battle. CRISPR is a potentially revolutionary technique that is so broadly useful across medicine, agriculture, and industry that it could earn the dispute’s winner billions of dollars. Lawyers for the University of California, Berkeley, and the Broad Institute of MIT and Harvard both argue that their scientists had invented CRISPR first.

The hearing was part of something called an interference proceeding—a piece of patent law that is quite literally archaic now. On March 16, 2013, the U.S. patent system switched the way patents are awarded: Previously, a patent was granted to the very first party to invent something; now, a patent simply goes to whoever files a patent application for an invention first. As it happened, Berkeley filed its initial patent just one day before the March 16 switchover. So here we are trying to figure out who invented CRISPR first.

CRISPR will be one of the last interference proceedings, and quite likely the last big, splashy one. The young woman in front of me in line was a junior patent examiner herself, curious to witness an interference proceeding before the whole thing went away. Another man was a patent lawyer, preparing for oral arguments in a far more obscure upcoming interference proceeding. Well-dressed lawyers for companies that had licensed the disputed patents were on hand to watch it play out, too.

The abbreviated version of the disputed history of CRISPR goes like this: In 2012, a team led by Berkeley’s Jennifer Doudna published a paper showing how to use CRISPR to edit DNA in a test tube. Six months later, another team led by the Broad’s Feng Zhang used CRISPR to edit mammalian cells, which he claims to have been working on even before Doudna’s paper came out. Concurrent with publishing scientific papers, Berkeley and the Broad were filing patents. Berkeley filed first, but the Broad paid for an expedited review and was awarded a CRISPR patent in April 2014.

The key legal question in the case is, put very simply, what is the actual invention that's being patented? Is it streamlining CRISPR to edit any DNA, period, as Berkeley argues? Or is it getting CRISPR to work in complex animal and plant cells—as opposed to floating DNA in a test tube—as the Broad argues?

The lawyers made their cases in front of a panel of three judges in a hearing room. But because so many people showed up to watch, they had to open a second hearing room for overflow. When the overflow room filled up, a moment of panic flicked over the two dozen or so people still unseated before the staff opened a third hearing room. This room still did not have enough chairs. A few people climbed up on the dais and sat behind seats usually reserved for judges.

Back in the first hearing room, lawyers on both sides had 20 minutes to make their oral arguments. The Broad’s lawyer went first, opening with—to give you a taste of the legalese—“With no substantial dispute over our motion three, exhibit one meets count one.”

Berkeley’s lawyer went next, and he got most of the tough questioning from the three-judge panel. The judges were keen to ask whether, as Berkeley argued, it was really so obvious that CRISPR would work in complex animal and plant cells after Doudna’s group had found it working in simple bacterial cells. If so, then Berkeley could claim that editing CRISPR to edit any DNA was patentable. If not, then it supported the Broad’s contention that the leap to plant and animal cells was the real invention.

But trying to divine the outcome of an interference proceeding from questions asked is no more advisable than trying to divine a Supreme Court ruling from questions during oral arguments. The patent office could make a decision in the next few months, or it could call a second hearing.

And even once the interference proceeding is over, one side could file for an appeal in the U.S. Court of Appeals for the Federal Circuit. The two sides have filed nearly 900 documents for the case already. Unless the case is settled—and lawyers do not think it will be—expect even more filings before it’s over.