Does UC’s new CRISPR-Cas9 patent really cover eukaryotes?

Zachary Silbersher

In the latest episode in the long-running CRISPR-Cas9 patent battle between the University of California and Broad, UC has obtained a new patent related CRISPR-Cas9.  UC has touted this patent, as well as another expected to issue shortly, as “useful to locate and edit genes in any setting, including within plant, animal, and human cells.”  So, did UC just win patents covering CRISPR-Cas9 in eukaryotes?  How does this square with the patent interference that UC recently lost at the Federal Circuit on this very issue?

UC’s new patent is U.S. Patent No. 10,226,611.  The first important thing to note is that the ‘611 patent is a continuation of U.S. Patent Application No. 13/842,859.  That is the same application that spawned the patents at issue in the prior interference.  This raises an immediate question.  If UC’s earlier patents spawning from the ‘859 application could not claim using CRISPR-Cas9 in eukaryotes, why is the new ‘611 patent any different?

As we previously discussed, on September 10, 2018, the Federal Circuit issued its decision reviewing the interference decision.  Following that, on September 26, 2018, UC disclosed that decision to the Patent Office during prosecution of the ‘611 patent.  Following that, the Patent Office issued a Notice of Allowance for the ‘611 patent on January 3, 2019.  So, from that, we know that the Patent Office was clearly aware of the Federal Circuit’s decision before allowing the ‘611 patent.

Indeed, in the Notice of Allowance, the Patent Office referenced the holding of the Federal Circuit, and quoted the Court’s statement, “the parties claim patentably distinct subject matter” because “the invention of such systems in eukaryotic cells would not have been obvious over the invention of CRISPR-Case9 systems in any environment, including in prokaryotic cells or in vitro, because one of ordinary skill in the art would not have reasonably expected a CRISPR-Case system to be successful in a eukaryotic environment.”   

Nevertheless, the Patent Office concluded that UC’s current ‘611 patent is patentable.  The Patent Office explained that it is “directed to the use of CRISPR-Cas9 in a ‘cell,’ which broadly encompasses any such cell, do[es] not interfere with claims that are directed to the use of CRISPR-Cas9 in a cell that is ‘eukaryotic’ specifically.”  (14/942,782, Notice of Allowance at 3).

This is an interesting interpretation by the Patent Office of the Federal Circuit’s prior interference decision.  To understand why, let’s take the history of this dispute step-by-step.   

Going back several years, UC filed patents claiming priority to the ‘859 application that are directed to CRISPR-Cas9 in any environment.  Broad then took that basic technology, and implemented CRISPR-Cas9 in the eukaryotic environment.  Then Broad tried to patent CRISPR-Cas9 within the eukaryotic environment (which is really the gold ring of this breakthrough technology.)  UC objected, and started the interference proceeding at the Patent Office.  In the interference, UC claimed that it had, in fact, invented implementing CRISPR-Cas9 in the eukaryotic environment.  It argued that using CRISPR-Cas9 in the eukaryotic environment was obvious from what was disclosed in its ‘859 application, which preceded Broad’s patents.

But both the Patent Office and the Federal Circuit rejected UC’s argument.  They held that implementing CRISPR-Cas9 in the eukaryotic environment was not obvious from UC’s ‘859 application.  The primary reason was because, based upon the evidence before these courts, scientists were unsure whether the system disclosed by UC in the ‘859 application would actually work within eukaryotes.  Indeed, this is what the Federal Circuit stated in its September 10, 2018 opinion: “one of ordinary skill in the art would not have reasonably expected a CRISPR-Case system to be successful in a eukaryotic environment.”   

Because of that, the Federal Circuit acknowledged that the UC’s discovery of CRISPR-Cas9 in vitro was patentably distinct from Broad’s alleged discovery of CRISPR-Cas9 in eukaryotes.  Yet, the Patent Office has now interpreted that finding by the Federal Circuit and applied it to UC’s instant ‘611 patent.  And the Patent Office has essentially found that UC’s new patent, which is directed to “any such cell” is patentably distinct from Broad’s patent, which is directed to “a cell that ‘eukaryotic specifically.”   

It is a bit difficult to reconcile the current logic of the Patent Office.  Technically, UC’s new patent, which is directed to “any cell,” may be patentably distinct from Broad’s patent covering eukaryotes specifically.  But eukaryotes are a species of a larger genus including “any cell.”  The key inquiry in the interference—which UC lost—was whether the ‘859 application rendered obvious using CRISPR-Cas9 in eukaryotes.  The Federal Circuit held it did not.  That is another way of saying, UC did not invent using CRISPR-Cas9 in eukaryotes—at least not in the ‘859 application.  And that is another way of saying, UC’s ‘859 application did not adequately describe using CRISPR-Cas9 in eukaryotes. 

And therein is the rub.  “Written description” is one of the measures of patentability.  The written description requirement is codified in 35 U.S.C. § 112.  When a patent claims an invention that is not adequately described in the application, the patent is typically invalidated for lack of written description.

Here, that very logic would appear to apply to UC’s new patent.  During the interference proceeding, UC strenuously argued that its ‘859 application showed that it invented using CRISPR-Cas9 in eukaryotes.  Yet, the Federal Circuit disagreed.  The Court’s holding essentially found that the ‘859 application did not invent or adequately describe using CRISPR-Cas9 in eukaryotes.  Because UC’s new ‘611 patent is a continuation of the ‘859 application, it would appear to be infected by the same issue.

 For reasons previously discussed, this author did not necessarily agree with the Federal Circuit’s analysis.  But be that as it may, the decision is what it is.  Unless and until that decision is reversed en banc or by the Supreme Court, it essentially remains the definitive holding on this issue. 

What does all of this mean?  It means that UC may have just procured a patent that technically covers using CRISPR-Cas9 in eukaryotes, but that patent is not worth much.  Why?  Because, if UC attempts to enforce its new patent against anyone for using CRISPR-Cas9 in eukaryotes, those accused infringers could easily invalidate it.  They would point out that the patent claims are broad enough to cover use of the technology in any cell, including eukaryotes, but the Federal Circuit has already—and rather definitively—concluded that that invention was not adequately disclosed in the ‘859 application, and correspondingly the ‘611 patent too. 

Ok, but what if UC has no intention of actually suing anyone?  What if UC only seeks to license this patent?  That may be.  But any prospective licensee would be able to negotiate down the price since the risk of not taking a license, and being sued by UC for infringement of the ‘611 patent, is not very risky.  In fact, the risk is fairly low if the Federal Circuit has already mapped out a nearly irrefutable way to invalidate the patent.

The greater risk for UC is that having its own patent, the ‘611 patent, that purportedly covers using CRISPR-Cas9 in eukaryotes, does very little to eviscerate the threat of Broad’s patent.  Broad could still sue UC—or any of its exclusive licensees—for infringement of its patents, and win.  Having your own patents does not insulate you from liability for infringing someone else’s patents.  Technically, UC could countersue with the ‘611 patent, but for all the reasons already discussed, the strength of that suit remains weak.

Well, maybe all of this is besides the point.  UC has undoubtedly been a trailblazer in this area, and its patent woes are—in one view—a questionable failure of the patent system.  But, be that as it may, its strategy may not be to necessarily get the best patents possible.  It may have already lost the opportunity to do that.  Rather, it seeks to collect as many patents as possible related to CRISPR-Cas9 for use in plant and animal sells, and build a small pile of patents that any start-up interested in this space would be wise to take a license to.  Sometimes the devil is in the details, and sometimes not.