The first part of this article is here.
The exclusive license that UCLA granted to Medivation meets the requirements for an assignment of the subject invention. The license grants substantially all rights in the subject invention (to all compounds for use in treating disease), grants the right to grant sublicenses (including exclusive sublicenses), and grants the right to sue for infringement. The license reserves educational and research uses, reserves rights to use for Howard Hughes Medical Institute (based on the inventive contribution of one of its employees working as a faculty member at UCLA), and reserves rights for the U.S. government. That is, courts have ruled that such a license instrument constitutes an assignment of the invention. This, too, is prohibited under Bayh-Dole, unless the federal government approves, and regardless, the assignee has to accept the nonprofit patent rights clause (see 37 CFR 401.14(a)(k)(1)).
What does it mean that Medivation, Montreal (Pfizer’s subsidiary), and Pfizer must accept the nonprofit patent rights clause that UCLA invented under? It means that each the new assignee must pass on the nonprofit patent rights clause in any subsequent assignment of the subject invention, must share royalties with the inventors quite apart from anyone else who is sharing royalties with inventors, must have a preference for small businesses in licensing the invention, and–and this is the poison pill–must use any balance after expenses incidental to administrating subject inventions for “scientific research or education.” If KEI wants to go after Xtandi, the finger down the throat of rights is right here, not in the march-in provisions. If UCLA has assigned the invention, then the assignee must use any income with respect to that invention for the public purposes of scientific research or education. Nothing else.
If Pfizer does not want to be exposed to the financial liability of allocating not only its profits but all of its income less those administration of subject invention expenses to public purposes, then it will cough up the assignment–of all substantial rights–like a hairball. No doubt Pfizer would be happy (in the circumstances, relatively speaking) with a non-exclusive license to make and use with an exclusive license to sell. Medical insurance companies might then fund companies to give away enzalutamide to their patients, so there would be no sale involved.
There’s more to it, here, too. In 2009, Medivation entered into a collaboration agreement with a Japanese pharmaceutical company, Astellas. Under the collaboration agreement, Astellas receives an exclusive sublicense to the UCLA patent for worldwide manufacturing and sales of RD 169/MDV3100/Xtandi. In the U.S., Astellas and Medivation share sales, agreeing to split costs 50/50. Why does this matter?
Bayh-Dole requires (35 USC 203) that for any exclusive license to sell or to use in the U.S., the licensee must agree to work with product “manufactured substantially” in the U.S. Apparently, however, Astellas manufactures Xtandi outside the U.S., but for limited manufacturing that Medivation contracted out to provide a supply of the drug for use in clinical trials. Astellas’s manufacturing plants are in Ireland and the Netherlands. It would appear that Xtandi is not being “manufactured substantially” in the U.S. for the U.S. market. This condition–if there is no U.S. manufacture–could support a march-in by the federal government. Again, however, Bayh-Dole is written to permit plenty of wiggle–any involved party (contractor, inventor, assignee, or exclusive licensee) can appeal and no march-in can take place until the appeal is exhausted. But Bayh-Dole makes this form of march-in even more difficult and unlikely:
However, in individual cases, the requirement for such an agreement may be waived by the Federal agency under whose funding agreement the invention was made upon a showing by the small business firm, nonprofit organization, or assignee that reasonable but unsuccessful efforts have been made to grant licenses on similar terms to potential licensees that would be likely to manufacture substantially in the United States or that under the circumstances domestic manufacture is not commercially feasible.
So the federal government has no obligation to march-in, even for the manufactured substantially in the U.S. requirement. It’s not a law–it’s more like guidelines, and more rather like a puffy statement of feel-good that adds administrative baggage but is never used–and it never has been used.
However, in the case of Xtandi, it appears that domestic manufacture is very “commercially feasible.” The problem is whether such manufacture also comes “under the circumstances” by which UCLA granted its exclusive license to Medivation. One would think Bayh-Dole could have been drafted to protect the public interest–if that was indeed the purpose of all this apparatus in the law. But it appears that Bayh-Dole was drafted to protect the interest of patent brokers and monopolist corporations. Bayh-Dole is a do WTF you want law. The economy no doubt receives some benefit from patent brokers and monopolists, too. But then why not come out and declare that as public policy. Drop the failed apparatus, the march-in rights, the puffy language about public availability on reasonable terms? Why not just demand the government’s license and leave it at that?
There is one more twist (of many) to discuss. We have already looked at the issue of nonuse–UCLA licenses many compounds into nonuse. And we have looked at exclusive licensing–UCLA assigns its invention to Medivation under the cover of an exclusive patent license. Bayh-Dole also requires nonprofits to use licensing income after recovery of expenses incidental to managing subject inventions for “scientific research or education.” But UCLA has also sold its future royalties to a Royalty Pharma, a company that buys out royalty streams. That is, UCLA has also sold its financial interest in the assignment to another company. But Bayh-Dole forbids nonprofit assignment of a subject invention unless the nonprofit patent rights clause is included.
Here, however, UCLA has assigned not only the invention but also the income from the deal. The income from the deal–what Medivation owes to UCLA under the exclusive license now gets transferred to another company. In exchange, UCLA gets a payment from Royalty Pharma that’s not income from licensing, but rather is income from betting on the future value of a license. In essence, Royalty Pharma is betting that UCLA has undervalued its future royalty stream, or is willing to undervalue that stream in exchange for a single up-front payment. That is, if Royalty Pharma is correct, it has purchased UCLA’s royalties for less than UCLA would have received–and therefore UCLA has used some portion of its royalties to purchase–its royalties. Or, another way, UCLA has shared its royalties with an intermediary investor, quite in defiance of UCLA’s obligation under the nonprofit patent rights clause, 37 CFR 401.14(a)(k)(3).
There are two problems with this. One is that this transaction is not allowed–it is not a use of licensing income for scientific research or education, but rather trading that income for other income. Laundering, as it were. Two is that UCLA has announced that it will “invest” the payment it receives. Again, purchase of stocks (or real estate) is not scientific research or education. Even if the proceeds from investment go to such things, the income itself is being used for investment, not for scientific research or education. Again, it would appear that this arrangement is a breach of the federal patent rights clause for nonprofits authorized by Bayh-Dole. UCLA has no authority to redirect its licensing income in exchange for a payment. The only way this is not a breach is if Royalty Pharma pays UCLA *more* than UCLA would have received without a buyout of the revenue stream. Furthermore, UCLA does not have the right to invest royalty income rather than spend the income on scientific research or education.
But these objections to UCLA’s dealings–nonuse of compounds, license that’s really an assignment, laundering royalties–don’t matter if Bayh-Dole is a law without an enforcement apparatus. In practice, federal agencies let university contractors such as UCLA do whatever they want. Bayh-Dole is a do WTF you want law. Now that’s interesting public policy.
We have looked at a number of alternatives to march-in based on price to deal with Xtandi. UCLA assigned a whole class of compounds to Medivation, which developed only one. The rest are subject to march-in based on nonuse, not price. Perhaps one or more of the other compounds are as effective against cancer as enzalutamide. There would be competition, then.
It is also clear that most of the compounds assigned to Medivation have not been developed or used. They are most obviously in a state of nonuse, behind the patent monopoly. It would appear to be a straight shot for NIH or DoD to march-in on all of these other compounds. They should be available for any other development or use. Make Pfizer cough them up. Not the immediate hairball of interest, but one that gets Pfizer used to the idea of coughing up rights held in violation of Bayh-Dole’s patent rights clause.
Same for march-in based on a failure to manufacture substantially in the United States–it’s just that Bayh-Dole gives federal agencies to waive the requirement. But it is also the case that it appears that the circumstances don’t meet Bayh-Dole’s conditions for such a waiver. If Pfizer doesn’t want to manufacture substantially in the United States, then it should insist that its license downgrade to non-exclusive in the United States. Otherwise, it appears to be exactly the situation that march-in was intended to address–the most important provision in Bayh-Dole by its own assertion, in a law that takes precedence over all other laws pertaining to federal policy for inventions made in federal contracting. One would think that 35 USC 203 march-in for failure to manufacture substantially in the United States would be something that the present administration would consider a priority. This is the case to make the point, if there ever was one. Again, however, march-in requires a willing federal agency, and historically the NIH has been more than willing to circumvent, misrepresent, and undermine executive branch patent policy and then even Bayh-Dole, which its own patent counsel drafted. Someone in the federal government has to have the courage to step up–or be required by the Department of Justice to do so.
UCLA, in assigning a subject invention, imposes the nonprofit patent rights clause on the assignee, now Pfizer. UCLA’s inventors have a cause of action against Pfizer for not sharing royalties with them. It does not matter that UCLA shared royalties with them from its assignment of their invention to Medivation. Pfizer has the same obligation as UCLA has under its nonprofit patent rights clause. Where’s the money? If UCLA’s inventors sued for a share, they would make it clear that Pfizer is an assignee. That makes things even more interesting.
As an assignee, Pfizer is limited in how it can use income with respect to a subject invention. It can deduct the costs incidental to the administration of subject inventions. It can share royalties with inventors (hasn’t so far). But anything left has to go to support scientific research or education. This is a very, very big number at this point. We are talking billions. Who has standing to enforce this federal regulation? UCLA? After all the clause is required to go along with UCLA’s action? NIH or DoD, the federal agencies involved? They are on the other side of the funding agreements with UCLA and the nonprofit patent rights clause. The Department of Justice? It would appear that this is a nice opportunity for a whistleblower–it’s just that it is the public that has been shorted the billions of dollars, not the federal government.
Finally, UCLA has disregarded the nonprofit patent rights clause by selling its future royalty income for a lump sum payment and investing the money instead of using it to support scientific research or education. Going after UCLA for compliance won’t change the assignment to Pfizer, but it would carry some little satisfaction that UCLA should comply with all elements of the nonprofit patent rights clause, not just the do WTF you want ones.
In all of this, it would appear that the hairball comes up not with march-in but with a demand for accounting for all of Medivation’s and Pfizer’s income with respect to the subject invention that was assigned to them under UCLA’s original exclusive license.