Painful Lesson on Patents
Giving someone something and agreeing to give someone something are two very different things -- and Stanford University officials, to their dismay, are probably more aware of that distinction now than they were just a few days ago.
On Wednesday, the U.S. Court of Appeals for the Federal Circuit told a lower court judge to dismiss a lawsuit Stanford had brought accusing the pharmaceutical company Roche of infringing its patent on a technology that measures the concentration of HIV in blood plasma. Stanford lost the case, essentially, because its policy on who owns inventions created using university resources required researchers, at some future date, to "agree to assign" ownership rights to the university; the comparable policy at Cetus, the Roche-owned company with which the Stanford researcher did outside work, in turn, said that an inventor "will assign and do hereby assign" his or her rights.
Because the agreement with the outside company took effect immediately at the point the invention developed, the researcher was unable to assign his ownership rights to Stanford, leaving the university in the cold, in the court's judgment. "[B]ecause Cetus’s legal title vested first, [the researcher] no longer retained his rights, negating his subsequent assignment to Stanford during patent prosecution."
Several experts on intellectual property law said the Stanford v. Roche case has significant implications for any of the many colleges and universities that seek to transform their faculty's scholarly work into patentable inventions. "Every institution should look at its own agreements with investigators, and know what its researchers are doing in collaborative arrangements" with outside companies or entities, said Ron Eisenstein, a lawyer at Nixon Peabody in Boston who works with numerous universities.
Permissive policies on intellectual property may help institutions attract researchers interested in monetizing their scholarly work as well as companies seeking to collaborate, Eisenstein and others said, but the Stanford case clearly shows the downside of such policies, too.
Stanford and Roche spent much of the last decade discussing, arguing and finally fighting in court over rights to the HIV test that two Stanford researchers had patented in the early 1990s. Stanford sued the company in 2005, after the university and Roche failed to agree on a licensing agreement, and a federal lower court judge issued a split decision in 2006, which both sides appealed. In its reconsideration of the case, issued Wednesday, the Court of Appeals for the Federal Circuit stripped away numerous ancillary issues and went back to the very beginning.
More than two decades ago, when one of the researchers in question, Mark Holodniy, signed his Copyright and Patent Agreement upon joining Stanford in 1988, the university's administrative guide on intellectual property proudly said: “Unlike industry and many other universities, Stanford’s invention rights policy allows all rights to remain with the inventor if possible.”
It's not surprising that universities and other research institutions would want to have patent policies that are attractive to inventors, said Michael C. Martensen, an intellectual property lawyer at Hogan and Hartson, in Colorado Springs. Many top researchers who choose to work in academe do so because they "enjoy the academic environment, but they're also usually thinking about how do I leverage this knowledge [for commercial purposes], and to attract that kind of talent, schools deliberately hang a carrot out there" that offers the promise of ownership, Martensen said.
Consistent with the university's stated philosophy at the time, the Federal Circuit court said, the patent agreement that Holodniy signed with Stanford said: "I agree to assign or confirm in writing to Stanford and/or Sponsors that right, title and interest in ... such inventions" created using contracts or grants that flowed through the institution.
"We have held that the contract language 'agree to assign' reflects a mere promise to assign rights in the future, not an immediate transfer of expectant interests," the appeals panel wrote. "Therefore ... Holodniy agreed only to assign his invention rights to Stanford at an undetermined time."
The "visitor confidentiality agreement" Holodniy signed when he began working with Cetus on an HIV treatment was much clearer: "I will assign and do hereby assign to Cetus, my right, title and interest in each of the ideas, inventions and improvements.... In contrast to the CPA, the [Cetus agreement]'s language of 'do hereby assign' effected a present assignment of Holodniy's future inventions to Cetus," the Federal Circuit opinion said. "Therefore, Cetus immediately gained equitable title to Holodniy's inventions."
All of the other arguments and maneuvering in which Stanford and Roche engaged in the patent infringement lawsuit, the appeals panel ruled, essentially pale before the simple fact that Holodniy gave Roche, and not Stanford, his rights to the HIV technology.
"Stanford would have won the case (at least the ownership issue) if it had written its employee agreement to automatically transfer rights [to the university] upon creation of the invention," Dennis Crouch, associate professor at the University of Missouri School of Law, wrote in a post on his patent law blog, Patently-O.
Scrutiny at Stanford
Patrick Dunkley, senior university counsel at Stanford, said that university officials were reviewing the decision and deciding whether to appeal it.
In terms of the language about assignment of rights on which the court focused, Dunkley said that Stanford officials had not paid significant attention to it previously because "in the grand scheme of the panoply of issues in the case, and there were many" -- interplay with the federal Bayh-Dole Act, arguments over the statutes of limitations -- "this did not seem like the issue on which the case would turn."
He questioned the court's suggestion that Stanford had purposefully taken a permissive stance in granting ownership rights to researchers, saying that the language the court cited about seeking to have "all rights ... remain with the inventor if possible” can no longer be found among Stanford's administrative policies. And Stanford's patent agreement with employees clearly said in 1988, as it does now, that researchers may "not enter into any agreement creating copyright or patent obligations in conflict with this agreement."
"The clear intention of our policy, as long as I've been here, is that the university owns any technology that is developed on Stanford’s time and through Stanford’s resources," Dunkley said. (The University of California's patent acknowledgment form -- which, it must be said, is included in the same document in which employees must swear their "oath of allegiance" to the state -- uses language similar to Stanford's, in which an employee "acknowledge[s] my obligation to assign inventions and patents that I conceive or develop while employed by University or during the course of my utilization of any University research facilities.")
Dunkley acknowledged, though, that in the court's view, the wording of its policy in 1988 did not make that intention clear. Stanford's current Patent and Copyright Agreement for employees, which was updated in 1995, contains slightly different language, but it still appears as if it would run into trouble using the court's analysis because it uses the "agree to assign" language: "I further agree to assign to Stanford all my right, title and interest in such patentable inventions and to execute and deliver all documents and do any and all things necessary and proper on my part to effect such assignment."
Legal experts said they suspected Stanford would have to adjust its current language further to meet the court's standard. "I suspect that what they’re going to have to do from here on out is to have all researchers say that they are 'hereby assigning all rights' to Stanford," said Martensen of Hogan & Hartson.
Bruce T. Wieder, a patents expert at Dow Lohnes, in Washington, said that if Stanford is intent on maintaining meaningful ownership rights for professors, it could alter its agreements to make clear that the university owns the invention from the start, but to "promise that, if at all possible, the rights will be assigned back to the inventor."
He said that given its experience in this situation, Stanford -- and perhaps other institutions -- should consider making it a habit to "see outside agreements" that researchers sign with companies. "If these are people are working for me, and I’m thinking I have rights to the stuff they’re working on, I really should know if I’m wrong," he said.
Dunkley said it would be impractical for Stanford or another large research university to review all agreements that researchers sign with other entities, since in many cases they may have to sign "visitor confidentiality agreements" at the reception desk of a large corporation they are visiting. ("If I walk in the lobby at Google, I have to sign one," he said.)
Still, Stanford is certainly likely to review its policies on a range of fronts in the wake of the appeals court's decision this week, Dunkley said. "Anytime one gets an opinion like this that, if it stands, would alter substantive rights such as this," he said, "it would only be prudent to go back and review any policy that might be impacted."
Search for Jobs