The American Prospect has an interesting piece (“A Big Miss on Drug Prices”) on the NIH yet again defying the Bayh-Dole Act and refusing to launch an investigation into whether UCLA and Pfizer have met Bayh-Dole’s standard (35 USC 201(f)) to make the benefits of using a subject invention available to the public on “reasonable” terms. As Dayan points out, the NIH in its rejection of a petition to start march in proceedings, happily leaves out the “reasonable terms” part of the 35 USC 203(a)(1) condition for march in.
NIH refuses not only not to march in but also not to even conduct an inquiry to ascertain whether UCLA and Pfizer have met the requirement of “on reasonable terms.” The “reasonable” in 35 USC 201(f), given the remedy is licensing to introduce more independent use of a given subject invention, carries an effective meaning of “competitive.” Reasonable terms are those that would be reasonably expected if there were competitive use of the subject invention. NIH just ignores the law and also ignores its responsibility to the public under the law. The effect of NIH’s refusal to even start a march in procedure to evaluate the situation surrounding Xtandi is to claim that Congress passed Bayh-Dole with public protections delegated to federal agencies but did not expect federal agencies to protect the public by using any of them.
In all of this, it is important to understand that the Bayh-Dole march in right is not the first line of public protection in Bayh-Dole. March in is, like, number three. Not that it appears to matter much to the NIH, which is not about to use any of the rights Congress secured for the federal government to protect the public from patent abuse, especially in matters of public health and safety.
Let’s review march in. March in is directed at patent exploitation in non-governmental markets–that is, where the government is not involved in making, using, or selling anything within the scope of a subject invention. Under the march in provisions–35 USC 203–Bayh-Dole distributes rights in subject inventions between a contractor and the federal government represented by a funding agency.
In the case of enzalutamide (sold as Xtandi), UCLA, under Bayh-Dole’s nonprofit patent rights clause, grants to the federal government certain rights in its subject inventions(s). For instance, the government gets from UCLA a royalty-free worldwide license to practice and have practiced the subject invention for or on behalf of the United States. UCLA also agrees that the government gets a limited right to determine the way in which UCLA exploits its patent rights in non-governmental markets–called “march in.” These rights are confirmed through a patent rights clause that is part of every “federal funding agreement.”
The patent rights clause is, in effect, a federal version of a technology administration agreement between joint owners of a patentable invention. March in is one provision that distributes certain rights–the contractor agrees to the distribution of march in rights in the management of subject inventions, along with other rights in each subject invention.
Universities do technology administration agreements routinely with other institutions when there are inventions with joint inventors at other institutions. One university grants the other the right to be lead on filing a patent application, the universities agree on how patenting costs will be handled, and the passive university usually retains a right to approve the other university’s patenting decisions. Technology administration agreements usually also distribute rights in inventions for licensing decisions. So one university takes the lead in licensing negotiations and license drafting, and the other has rights to review and conditionally object to the lead university’s licensing plans and actions. Getting these arrangements in place is especially important for exclusive licensing, assignments, and patent enforcement (both of which generally require all owners of an invention).
Bayh-Dole’s march in provisions create a technology administration agreement between the holder of title to a subject invention and the federal government–the government’s rights are invention rights. In some cases, such as failure to fully and timely disclose a subject invention to the federal government, the contractor gives the government the right to take title to the invention–and therefore have the right to decide about patenting–subject to the contractor receiving back a non-exclusive license to the invention. If the title holder (or its assignee or exclusive licensee–call them “ilk”) does not make product based on the use of a subject invention available on reasonable–or competitive–terms, then the government can request that the title holder (or ilk) grant licenses, or more licenses, to introduce competition (that’s the “free competition and enterprise” identified in Bayh-Dole’s statement of policy and objective at 35 USC 200).
The Bayh-Dole patent rights clauses are technology administration agreements. The federal government is almost but not quite given status as a joint owner of each subject invention. Instead, title remains with the contractor–such as UCLA–but then the contractor has to grant to the federal government some of the rights of a joint owner (and more, but conditionally). For instance, the right to receive a full disclosure is a condition of leaving title with the contractor. No full disclosure, no timely disclosure, and the federal agency has the right (the contractor gives the government this right as a condition of the funding agreement) to request title. And the government’s right to practice and have practiced a subject invention is conditioned on the contractor electing to retain title in the subject invention.
There’s no “extinguishing” a patent on a subject invention through march in, as The American Prospect describes it. Bayh-Dole requires contractors to grant to the federal government the conditional right to request that the contractor (or assignee or exclusive licensee) grant licensees. If the contractor (or its ilk) grants the licenses, then the contractor (or ilk) also gains the benefit of compensation for those licenses. The patent is still in play, but the government has requested–and the contractor has agreed that the government has this right to request–that the contractor (or ilk) exploit the patent in a different way, one that introduces “free competition” in situations in which the contractor has failed to ensure products (“use with benefits”) have been made available to the public on competitive (“reasonable”) terms. If the contractor (or ilk) refuses to grant licenses when the federal government makes the request, then the contractor has agreed that the federal government the right to do so and doesn’t owe the contractor anything for those deals. The contractor’s own deals (if any) are still in place–so the contractor’s patent stays in play. It’s just that those deals might earn income at a different rate if sales shift to competitively priced product. There’s still a patent–but march in changes how a contractor (or ilk) can be compensated for exploiting the patent.
March in then does not destroy a patent but limits how a contractor can exploit that patent. Patents on subject inventions are not ordinary patents. They are limited rights patents, and one of the key limitations is that contractors must share invention rights with the federal government.
March in requests change how a contractor (or ilk) receives compensation for holding a patent on a subject invention. It’s sort of interesting. If someone infringes a patent and the patent holder seeks to enforce that patent, a court decides what constitutes reasonable compensation for the infringing action. The same is true under 24 USC 1498 in cases where the federal government uses a patented invention without a license. The issue is what’s the “reasonable and entire” compensation for the use covered by the patent. In the case of “unreasonable terms”–such as an unreasonable price–the government does not dictate price but rather requests that either the price is competitive without competition, or there will be licensing to create competition, which in turn should lead to competitive prices. It’s then up to the contractor (or its ilk) to decide whether they want to be compensated at competitive prices, or not compensated at all when the government grants licenses. It’s the contractor’s choice, and the contractor agrees to it up front, as a condition of accepting the federal award. This is not extinguishing the patent, voiding the patent, seizing the patent. This is not eminent domain taking. This is a deal made up front, before federal money is spent.
It’s then consternating that the NIH rejects their delegated role in the deal–and apparently cannot even grasp the deal. A law too complicated, or a federal agency deliberately too stupid, or too compromised, to understand it.
There are better alternatives in Bayh-Dole to march in, especially for Xtandi. One of these is the government license to practice and have practiced, which the contractor is required to grant as a condition of electing to retain title to a subject invention at 35 USC 202(c)(4). Under the Kennedy executive branch patent policy, from which Bayh-Dole draws its language, “to practice” means “to make, to use, and to sell.” The government license divides rights in a subject invention into a governmental market and a non-governmental market. March in covers the non-governmental market. The government license to practice and have practiced covers the governmental market. Xtandi straddles both markets. So we should at least take care of the governmental market. Think of the Xtandi needed by the Veteran’s Administration and by Medicare. These government organizations already have, as required by Bayh-Dole, an outright license to authorize generic versions of Xtandi for their “practice.” There are no conditions once a contractor has elected to retain title. There are no procedures or notice and there’s no provision for appeal–the license is irrevocable. The limitation is scope–“for or on behalf of the United States.” Straightforward. The government license does not even need the NIH to futz around first to decide whether to accept the license.
In Piscopo, then.
Government license!
No red tape!
Use it today!
Now!
The VA and CMS can authorize generic versions of Xtandi and have these generic versions sold to anyone within the scope of federal government or state government (“for or on behalf of the United States”) authority to act.
Yes, it’s true that the governmental market does not cover all users of Xtandi. But so what? It covers many of them. It makes all sorts of sense for the federal government to start with an action that is immediately within reach. The government stands to save millions of dollars a year by using its license to practice and have practiced. And that’s something. Furthermore, by authorizing generic versions of enzulutamide (Xtandi), the government prepares generic manufacturers to expand immediately when the UCLA patents expire. Finally, the government market takes a big slice out of the profits Pfizer obtains by selling into the government market at non-competitive rates–because it uses its patent to suppress competitors who would manufacture for the United States on competitive terms. These are very good things, and the government should act to make them happen.
A second line of action in Bayh-Dole to protect the public from patent abuse such as non-competitive pricing focuses on nonprofit disposition of rights in subject inventions, and this protection is also applicable to Xtandi. Under Bayh-Dole, nonprofits can assign subject inventions only if they also require the assignee to operate under the nonprofit’s patent rights clause. That is, the assignee becomes a party to the nonprofit’s funding agreement. The assignee becomes, according the definitions of Bayh-Dole, a new contractor–and must behave as a nonprofit contractor at that.
Why does this matter? First, an invention may be assigned by a formal assignment instrument that expressly conveys title. But an invention also may be assigned implicitly by an exclusive license to all substantial rights in an invention–the right to make, to use, and to sell. Universities routinely assign inventions under the cover of an exclusive patent license. They insist on retaining title to patents involved, but not title to the inventions involved. Very clever, eh? It’s a standard line that a written instrument is what it does, not how it is labeled. So, yes, an exclusive patent license may also convey ownership of the underlying invention. Any time a university grants an exclusive patent license for all substantive rights (plus, for the term of the patent, with a right to enforce), you are looking at an assignment of the invention. See, say, Ciba-Geigy v Alza for another, uh, clever way that UCLA exploited patent rights. And UCLA assigned the compounds from which Xtandi has been developed. Federal law, then–Bayh-Dole–requires Pfizer to comply with same nonprofit patent rights clause that applies to UCLA.
Now for the second reason all this matters. Under the nonprofit patent rights clause, a nonprofit must use all income earned with respect to a patent, less the costs of administrating subject inventions, for scientific research or education. See 35 USC 202(c)(7)(A). That same clause applies to the $20B or so that Pfizer (and Astellas) have earned (technically, the nonprofit patent rights clause applied to Medivation, which received UCLA’s exclusive license cum assignment, and Medivation then licensed non-US rights exclusively (also an assignment, apparently) to Astellas, and so there too the nonprofit patent rights clause applies). You see, if Bayh-Dole were enforced, then Pfizer would have to give up its sales income in Xtandi to scientific research or education. That’s not money that can go to shareholders, or to building new offices, or bonuses for executives. That’s big–and by NIH notions, too big to enforce.
But think about it. What does this restriction on assignments mean? It means that nonprofits are encouraged to license less than all substantial rights, even if they license some rights exclusively, and if they license all substantial rights, they must do so for significantly less than the term of any patent–and then license non-exclusively. Or, put another way, only the most charitable companies would ever think to accept an exclusive license from a nonprofit for all substantial rights. They would insist on a license of lesser scope, and in that they would accept a degree of competition from the start, or soon after the start, and in doing so they would avoid accepting the nonprofit patent rights clause and its restrictions on the use of income earned with respect to a subject invention.
Bayh-Dole then uses 35 USC 202(c)(7)(A) to promote its policy and objective of using the patent system to “promote free competition and enterprise.” That policy is not fluff. It is a limitation on ordinary patent rights, a limitation on how patent rights may be exploited for compensation–not by trolling, not by monopoly pricing, not by suppressing research and development, or by limiting availability because they don’t want to invest extra in production to meet the public’s need. If a company takes assignment of a nonprofit’s subject invention, the company accepts that whatever money it earns with respect to that subject invention will go to scientific research or education. Sure, it can turn itself into a scientific research and education company and pocket the income, but most companies are not going to do that. And most companies, therefore, including Pfizer, should reject any nonprofit’s offer to accept all substantial rights in a subject invention.
Congress could address this situation by authorizing an audit by the Office of the Inspector General for Pfizer’s compliance with the nonprofit patent rights clause. If Pfizer is not compliant, then, well, there’s billions to pay back. That would be interesting. That would be, say, statutory karma.
Now, if the federal government took the step of even hinting at either acting on its license to practice and have practiced or auditing Pfizer for compliance with the nonprofit patent rights clause, it would be reasonable to expect that Pfizer might actually come to prefer march in, maybe even insist on it. March in, it turns out, is the most favorable of these three options for Pfizer.
Meanwhile, and this will take many months, Congress must amend Bayh-Dole to take responsibility and enforcement of the patent rights clauses from the federal agencies and NIST. Clearly, funding agencies are compromised by the organizations that they attract with their funding. The relationship between funding agencies and contractors is too tight to also support a federal agency oversight function for inventions. Before Bayh-Dole there were draft bills on federal patent policy that proposed that oversight should be with a federal entity that did not fund research or development. DOJ comes to mind. That would be a good amendment.
A second action of Congress would be to require federal agencies to comply with Bayh-Dole rather than re-interpret the law to suit themselves. Perhaps there should be penalties for federal agencies that ignored Bayh-Dole requirements (most of the agencies ignore everything in Bayh-Dole but the administrative red tape).
A third action to pursue would be to amend federal patent law to preclude enforcement of patents for medical activities–essentially, roll back the carve out made by the pharma/biotech lobby in 35 USC 287(c). That would make pharma hopping mad, but then hopping madness (or withdrawing campaign donations) over losing some patent monopoly exploitation of the sick is not a particularly relevant argument when it comes to public health supported by public resources for public benefit.