Undermining Bayh-Dole by relying on it? 5

Now for Reimers’s new op/ed, making more arguments against march-in, but now directed at Xtandi, a prostate cancer drug based on a series of compounds developed at UCLA with federal funding. According to Knowledge Ecology International, Xtandi is offered for sale in the U.S. at a price of around $150,000 for a year’s worth of pills, more than double the price for which Xtandi is offered in other developed countries. Two generic versions of the drug already have FDA approval but cannot be sold in the U.S. because of UCLA’s patents and its exclusive license deal. The generic drug companies say they can make a profit selling their version of the drug for $5,000 for a year of pills.

In 2018, I wrote a series of articles about Xtandi and Bayh-Dole. The first one is here. The short of it is, UCLA violated Bayh-Dole in a number of ways to do its deal, and the outrageous pricing is one consequence. It’s not just that the federal government should march-in to address the pricing issue, or even that the government should use its license to make, use, and sell under Bayh-Dole and skip march-in for now–it’s that the government should enforce the other parts of Bayh-Dole that in their breach created the anti-competitive situation in the first place.

Reimers does not address this core point–that US prostate cancer patients get charged way more than patients in any other country, and a whopping 30x more than the price would be if there were competition to provide the drug. One might be left with the impression that if these weren’t “unreasonable terms” for U.S. cancer patients, then it is difficult to imagine any terms that would be “unreasonable.” If that were the case, then one also would have to accept that Congress in using “unreasonable terms” did not intend for there to ever be any unreasonable terms and so unreasonable terms must not mean anything, really, at all. That’s not a good starting point for statutory construction, but it is helpful to make the connection–arguments about Bayh-Dole carry with them implicit claims with regard to what Congress intended in passing Bayh-Dole. Not what Norman Latker or Senators Bayh and Dole privately thought they were doing, not even what Niels Reimers, who lobbied for the law, thought he was doing. What did Congress intend? What would one think is the plain and ordinary language of the law? Stuff like that. Work with it. Rule of law and all that.

The government contributed approximately $500,000 to research at UCLA that served as the foundation for Xtandi. The university’s findings were eventually licensed by Astellas — which, after more than $1.4 billion of investments and years of research and development, created the medication.

Garbled history. In 2005, UCLA licensed a series of compounds, including the one that became Xtandi, to an investment company, Medivation. Medivation outsourced most of its development work. It had one drug under development (which failed eventually in clinical trials). Enzalutamide, the compound on which Xtandi is based, became Medivation’s second drug candidate. Medivation then sublicensed rights to Xtandi to Astellas Pharma, a Japanese company. Astellas got exclusive worldwide rights, but in the US Medivation and Astellas both sell, and share costs equally.

Together, the two companies moved Xtandi through FDA approvals. As far as I can tell from public records, Medivation spent no more than about $300M developing Xtandi.

Prior to 1980, America had an innovation problem. The government funded basic research at university and non-profit labs and retained the patents resulting from the research. Inventors had no incentive to move their discoveries from the laboratory into the marketplace. It was up to the government to license these patents for commercial development.

Reimers used the same deceptive argument in his earlier op/ed. We have seen that the NIH and NSF operated IPA programs that permitted universities (and their patent management organizations) to acquire and retain ownership of inventions made in federally funded work. Whatever innovation problem America had before 1980, that problem did not have to do with federally funded basic research at universities. Just didn’t. Furthermore, university inventors had any number of “incentives” to move their discoveries from lab to marketplace, and they did so all the time. They started companies, they consulted with industry, they published as broadly as they could. They blew patent rights when it suited them, and they sought patents when they thought patenting would help. The incentives? Fame, public benefit, positioning to get more research funding, personal drive to make a difference, and even money. What Reimers wants his readers to believe is that federal funding was the only funding, and that researchers are motivate only by the money that might come to them by a share of university licensing revenue.

Read Reimers’s claim again:

Inventors had no incentive to move their discoveries from the laboratory into the marketplace.

It’s just nuts. I’ve worked with university faculty who could make more money in three months giving workshops on what they had invented than they would ever make from a  bit of royalties from a university patent license that might take five years or more to land, and years after that to result in a product for sale–if anything ever was licensed, which was rare.

Even if a federal agency did take ownership of an invention made in federally funded research, the inventor was assured of continued free access to the invention, since the federal government generally did not require licensing and even when it did, the license was generally royalty-free. With access, inventors (and other researchers) had incentives to work with companies to develop (improve, more inventing in the direction of applications) uses, standards, and products. Whatever proprietary positions were to be had–and with them, prospectively, ways to manage an emerging market–would come from doing more creative things–not just sitting around. Yes, these ways of making money might involve being hired, or working for shares in a startup, and not by having a bit share in whatever their university might receive by way of licensing income. But equity in a startup can scale just as nicely as a share of licensing income. Reimers’s assertion simply isn’t true. It would be weird to accept it as true–to do so would be to deride university researchers, which after a fashion, is what Reimers is doing here.

The process was a mess. Twenty-six different licensing policies governed the federal agencies funding research. The government often offered only non-exclusive licenses. Private companies were loath to invest their time and money in such a dysfunctional system.

There was mess, but it wan’t the federal agencies that were up to it. The federal agencies had their own policies, but most of them, and the various federal statutes, repeated the same consistent theme–if you were a company with a real market, then you could keep your inventions, except in special circumstances, and if you were not a company with a real market, then you could keep your inventions if you could make a case that keeping your inventions better served the public than would federal open access. In those special circumstances–set out in Congressional funding allocations and in the Kennedy and Nixon executive branch patent policies–the government would take ownership of inventions, may or may not file patent applications, but in either case would make the inventions available to all, mostly without even the formality of a licensing document because–well, what’s the point of a licensing document if the government has a mandate to make things available anyway.

Yes, there were different federal policies. They all did pretty much the same thing. They weren’t a problem except to university-affiliated patent administrators who wanted to take ownership of inventions but might get crossed up if the research had multiple federal agencies providing funding and one agency might be willing to release its claim on an invention while another agency might not. As far as I can tell, that possibility was a rare thing–if it ever actually happened. Let’s not forget, too, that every company that showed up at a university to sponsor research had its own expectations for how inventions, data, publications, and (later) software would be handled. University grants administrators and university-affiliated patent administrators had to deal with all that variation as well. Adding various federal agencies–with pretty consistent policies overall–to the mix is a non-issue. At best, it’s a matter of administrative convenience, which as important as convenience is to administrators, is usually at odds with anything inventive, innovative, in need of, say, opportunism.

Let’s pick up Reimers’s last claim again:

Private companies were loath to invest their time and money in such a dysfunctional system.

This is fake history. I’ve been through a lot of documents from 1950 to 1980 having to do with university research, technology transfer, patent licensing, and federal funding. I don’t see companies being loath or that the system (whatever Reimers intends that to be) was dysfunctional merely because federal agencies had flexibility with regard to whether their contractors could keep and patent inventions made under federal contract.

There is one area in which federal agency policy caused companies to step away–and that was in medicinal chemistry, in the early 1960s, and the reason did not have to do with patents or non-exclusive access but with a thing the companies called “contamination.” The Public Health Service made a policy change that insisted that if a company screened a compound discovered in work funded by the PHS, then the company in effect joined the public effort to develop the compound as a therapeutic and any other assets developed by the company on its own dime, as it were, were necessarily contributed to the public effort. Thus, if a company “participated” in the federal work, even if it screened compounds at its own expense, it might find that work it was doing on the same medical condition in parallel, on its own, would also be subject to a federal claim to ownership or to a non-exclusive government license. Contamination by public purpose. That could indeed be something to loath at a company. Public purposes. Bah!

So pharma companies boycotted federal research projects for a couple of years, the PHS backed off its policy change, and things went back to pretty much working.

In areas of engineering, companies had even fewer problems. Big ticket stuff like getting rockets to the moon and developing an internet were supported by companies working with the federal government and rather than stalking off all in a huff because they couldn’t patent themselves into being the sole source for all government technology, or exploit federal funding to gain patents to suppress all competition, they instead avoided most patents, created working standards, shared pre-competitive platforms, and competed on price, production, features, and non-essential claims. They made lots of money, anyway. Seems to have worked for them.

As a result, fewer than 5% of federally funded discoveries were licensed for commercial development. More than 28,000 taxpayer-funded insights languished. Not a single new drug was developed from federally funded R&D.

We have been through this claim. Not true. Deceptive. Nothing indicates the inventions covered by federally owned inventions “languished” or if they did it was because they were patented and not licensed exclusively. And the single new drug bit is untrue. Cisplatin and Flurourouricil are examples of drugs developed from federally funded R&D. Reimers is repeating a political talking point that has never been true. But remember, politicians are not expected to tell the truth, so, as the logic goes, they cannot lie because no-one should be expected to believe them.

The result opened a floodgate of American innovation. The law has helped grow the U.S. economy by up to $1.7 trillion. Over the past 25 years, U.S. universities and research institutions have been granted more than 80,000 patents, and some 70% of university innovations have been licensed to small companies.

It is difficult to find evidence to support Reimers’s claim about a “floodgate” related to Bayh-Dole. There is no indication that university patenting changed–it was growing before Bayh-Dole and kept growing at about the same pace after Bayh-Dole.

Reimers then cites some statistics. The first–“up to $1.7 trillion”–is the upper end of an unvalidated, pretty bogus economic model created with funding from AUTM, an organization of university patent administrators, that covered 20 years of licensing. Basically, they added up all the royalties reported to have been received by universities during those two decades, multiplied the result by 50 (making the sketchy assumption that all the reported income came in the form of a 2% running royalties on sales–and that there were even sales–or some other big number if they decided the royalty rate was 1% rather than 2%–it’s just a model so they can fiddle as they please), and then multiplying by how much that already big number might “contribute” to the economy. Even as a big number covering 20 years, it’s a really tiny number in the context of a multi-trillion dollar per year economy. And it’s an entirely made-up number anyway. Can’t say Reimers’s is lying because, well, why should anyone believe him?

It may be true that universities have got 80,000 patents in the past 25 years. In the Bayh-Dole era, they have got upwards of 120,000 patents, some 50,000 of which cite federal funding. But wait. Reimers is talking Bayh-Dole, but he is now citing statistics that don’t differentiate between inventions made in federal work and all other inventions. Turns out that universities are preferentially getting patents on non-federal inventions. Sixty percent of their funding is federal, but only about 45% of their patents. How odd. Floodgate, bah.

As for the 70% licensed to small companies, Reimers is off. Even at Stanford, the TLO licenses only about 20% of their inventions. So the 70% figure must be the percentage of those things that are ever licensed. But there’s more to Reimers’s problem. The only national survey of licensing is done by AUTM–those university patent administrators–and those folks count any license of $1,000 or more as if it were a patent commercialization license even if there’s no patent involved (software, biomaterials) and even if the purpose of the license isn’t commercialization but just buying access. Thus, one piece of code can be downloaded (with a $1,000 license) by 100 small companies and wowzers, it sure looks like the patent licensing program is preferring small companies.

Bayh-Dole doesn’t really care. Its primary standard of success is the utilization of inventions arising in federally funded research. Not the number of patents. Patents–by a count–is a measure of how much stuff has been excluded from public access. Until there’s some demonstration of how this exclusion has promoted utilization, the number of patents is a measure of what’s been stifled. Not the number of licenses, or whether the licenses are to small companies as a percentage of all licenses. Licenses just mean that a rights holder has promised not to sue to prevent another from practicing an invention. The core of Bayh-Dole practice is that a patent is not worth having unless it is to be licensed exclusively. That’s a lousy core principle any way, but that’s the one that university patent administrators make, that’s the one that the patent attorneys lobbying for Bayh-Dole made, and that’s the one that Reimers also has made. It’s implied in his op/eds, too. If Reimers wanted to persuade us, he would identify those inventions made in federal work that a university has patented, and for each tally up the date of first commercial use or public availability, and whether that commercial use was achieved with an exclusive license. Then we would have a basis to evaluate Bayh-Dole practice relative to non-Bayh-Dole practice for all the other inventions universities take under management.

But we don’t have that information. Reimers doesn’t have that information. AUTM does not see that information as worth bothering about. Neither do the federal agencies, even though Bayh-Dole’s standard patent rights clause suggests that federal agencies might ask for that information each year from each federal contractor electing to retain ownership of a subject invention. Bayh-Dole was amended in 1984, three years after it went into effect, to make all information a federal agency might collect exempt from federal public disclosure law. Even if federal agencies asked for the fundamental information that might show how Bayh-Dole was doing, it would be a state secret and we wouldn’t know. But one would think that if Bayh-Dole were actually doing really well, folks would gather real information and report it rather than substituting irrelevant and fake information and expecting any of us to believe them.

The Bayh-Dole Act contains a “march-in” clause that allows the government to require the patent owner to license additional companies if efforts are not being made to develop the technology.

The “if” clause here is sort of true, half-true, misleadingly untrue, and in its op/ed-y way, dishonest. Reimers claims that the present Xtandi march-in petition “grossly distorts” Bayh-Dole. But it is Reimers who distorts Bayh-Dole. The march-in clause is at 35 USC 203(a). There are four conditions under which a federal agency should march-in. Here’s the first, at 203(a)(1):

(1) action is necessary because the contractor or assignee has not taken, or is not expected to take within a reasonable time, effective steps to achieve practical application of the subject invention in such field of use;

There is plenty of looseness here. “Is not expected to take”–who gets to do the not expecting? “Within a reasonable time”–how long is a reasonable time? In the Kennedy patent policy, it was three years from the date of patent issuance, or about six years from the time of invention. The Harbridge House report in 1968 found that when companies owned the invention they had made, they had a high % of utilization within that time frame.

We have mentioned “practical application” in earlier articles in this series. Here’s Bayh-Dole’s definition of “practical application” at 35 USC 201(f):

The term “practical application” means to manufacture in the case of a composition or product, to practice in the case of a process or method, or to operate in the case of a machine or system; and, in each case, under such conditions as to establish that the invention is being utilized and that its benefits are to the extent permitted by law or Government regulations available to the public on reasonable terms.

Reduce it to the essentials:

The term “practical application” means [to work the invention so as]. . . to establish that the invention is being utilized and that its benefits are . . . available to the public on reasonable terms.

The heart of the definition of practical application–the part that is not evident from the words “practical application” themselves–is that the benefits of working the invention must be made (i) available to the public (ii) on reasonable terms. Not just “available” and not on just any “terms”–to the public, reasonable terms. To make benefits available to the public, it is reasonable to expect that one is not making the benefits available to some members of the public and not others, or granting access selectively, or discriminating against some members of the public, such as. not serving a given region or giving some members of the public special terms not available to all. To offer benefits on reasonable terms, and not just on any terms one wants, would appear to still leave much open to interpretation. Certainly, context matters. A high price for a medical intervention that’s expensive to produce, distribute, and support might be reasonable, while a lower price for a medical intervention that is super cheap to produce and distribute and support might still be an unreasonable price. There are a few standards one might choose, however:

(1) are the benefits affordable to most members of the public? Or, simply, do members of the public see the cost of those benefits as affordable? Yes. Just ask the public. For Xtandi, ask prostate cancer patients. I doubt many will say, yes, of course, $160,000 is fine for a year of Xtandi, especially since it costs only $5,000 or so to produce, and that with a decent profit.” No, they won’t say that. Even with insurance companies picking up $150,000 each year and leaving someone with a $10,000 bill for pills might still not be affordable for many, if not most, members of the public.

Even if someone argued that Astellas and Pfizer and Medivation spent $2B to test Xtandi, the public might point out that all these folks long ago recovered everything they had spent and more.

And if someone argued that Astellas and Pfizer would use the extra profit to invest in finding more drugs to develop, a reasonable public response would be that it’s unreasonable to charge prostate cancer patients a super high, even unaffordable price so the companies can do research on something else. Who nominated prostate cancer patients to bankroll a search for another profit-making drug? No, even the “we need that pile of money for research so it’s reasonable to get it from prostate cancer patients if we can” argument is unreasonable.

Maybe it would be reasonable for the companies to set aside some money to deal with adverse side effects, or to work on improved versions of Xtandi, and the like. Baking that into their margins might be a reasonable thing to do.

(2) A reasonable price is what a company would charge if there were multiple suppliers competing for business. In this definition, the public would find a price reasonable if the market was shaped by free competition–no backroom deals, no price fixing, no admonitions not to “race to the bottom” and wreck the clever scheme by which the public can be induced to pay, um, unreasonable prices. No industry front group lobbying to explain to the public that the crazy pricing is really just the price for having a pharmaceutical industry at all–that the companies involved would refuse to make any more medicines if they were told that they could not price gouge or discriminate on price.

Competition, so the theory goes, is a primary way by which a market establishes prices. In the case of “generic” drugs–drugs sold after the patents have expired–where there are multiple sources for these drugs, the price is often 1/10th that of the price for the drug when it was still covered by patents. A reasonable price, then, for a drug still on patent, but but based on an invention made in federally funded work, would the price that the drug would sell for if there were competition even when there isn’t.

In a pragmatic way, it is possible to see both of these possibilities for “reasonable terms” involving price or price discrimination working together. A significant portion of the public sees the price of a drug developed from an invention made in federally funded work as unaffordable. That perception should be sufficient to initiate a march-in proceeding. That does not mean that the march-in will result in a requirement for compulsory licensing, but it does mean that the contractor or assignee does have to provide an accounting to show that despite the high price, the costs of production and the like are also high. If the contractor or assignee cannot make a case for the high price being “reasonable,” then Bayh-Dole directs the federal agency to require licensing (or additional licensing). That is, Bayh-Dole directs the federal agency to restore competition, and the price–absent price-fixing and the like–should drop to a “reasonable” level, even an affordable level for most of the public.

You can see how “what the price would be if there were competition” does not dictate prices–it dictates competition if the price is unreasonably higher than it would be if there were free competition. You can also decide for yourself whether this line of interpretation somehow “grossly distorts” Bayh-Dole. I don’t see how it can, but then life is for learning.

Reimers, again, limits march-in to one situation:

if efforts are not being made to develop the technology.

Yes, that is the “nonuse” part of 203(a)(1) march-in. That’s the

the contractor or assignee has not taken, or is not expected to take within a reasonable time, effective steps to achieve . . .

If a contractor or assignee has not taken any steps, or the steps haven’t been effective, and some mysterious noun-thing does not expect effective steps to be taken “within a reasonable time,” then, yes, yes, that is a nonuse march-in condition. But the definition of “practical application” is broader than merely that there are “efforts.” Practical application is not that there are efforts–it is that there is use with benefits available to the public on reasonable terms. Once there is a product on the market, the development part of the effort is satisfied. What remains is whether the benefits of use are available on reasonable terms. It’s a gross distortion of Bayh-Dole to cite one instance among many and narrow march-in to that one instance, making it appear the other instances don’t matter.

Opponents of march-in being part of Bayh-Dole do the same sort of trick with “reasonable terms.” Whatever those terms might be, so the opponents go, they don’t include “price.” It’s utterly strange, given that if you ask most anyone waiting in line at your local pharmacy what “terms” they care most about with regard to getting their prescriptions filled, they will most certainly specify “price.” What other terms would they care about? Are they going to complain that the pharmacy has too much disclaimer language? Is that the “term” they will be concerned with? That the pills are too big or the wrong color or can’t be cut in half easily? No, it will be price, don’t you think?

It’s true. Bayh-Dole has “reasonable terms” and not “reasonable terms, including but not limited to, for the avoidance of doubt, a reasonable price.” By using “reasonable terms,” Congress indicates that its concern is broader than price, not that price is never a condition upon which a federal agency may march-in because Congress didn’t expressly mention price. That’s just dumb. Given that price is almost always front and center in any offering of a product or service to the public, if Congress had intended “reasonable terms” not to include price, Congress would have had to make that intention express.

The Supreme Court reasoned just this way in Stanford v Roche with regard to the meaning of “retain” and “of the contractor” in reviewing arguments for the interpretation of Bayh-Dole’s provisions regarding ownership of subject inventions.  Let’s look. First, on “of the contractor” (I’ve omitted citations):

Stanford asserts that the phrase “invention of the contractor” in this provision “is naturally read to include all inventions made by the contractor’s employees with the aid of federal funding.”

That reading assumes that Congress subtly set aside two centuries of patent law in a statutory definition. It also renders the phrase “of the contractor” superfluous. If the phrase “of the contractor” were deleted from the definition of “subject invention,” the definition would cover “any invention . . . conceived or first actually reduced to practice in the performance of work under a funding agreement.” Reading “of the contractor” to mean “all inventions made by the contractor’s employees with the aid of federal funding,” as Stanford would, adds nothing that is not already in the definition, since the definition already covers inventions made under the funding agreement. That is contrary to our general “reluctan[ce] to treat statutory terms as surplusage.”

Is the Supreme Court here “grossly distorting” Bayh-Dole? After all, scores of university attorneys signed on to amicus briefs arguing that Bayh-Dole vested ownership of inventions made in federally funded work with the universities (or gave the universities a first option to own, or a right of first refusal, or at least a mandate to take ownership–they never could get their act together on what exactly Bayh-Dole did, and maybe that was because what they argued for wasn’t in Bayh-Dole. They had just made it up.

The Supreme Court then offers a free lesson on reading federal statutes:

Construing the phrase to refer instead to a particular category of inventions conceived or reduced to practice under a funding agreement—inventions “of the contractor,” that is, those owned by or belonging to the contractor—makes the phrase meaningful in the statutory definition. And “invention owned by the contractor” or “invention belonging to the contractor” are natural readings of the phrase “invention of the contractor.”

Read the statute. Apply the plain and ordinary meaning of words. Does that result in a reasonable interpretation? There you go. Now for “retain”:

The Bayh-Dole Act’s provision stating that contractors may “elect to retain title” confirms that the Act does not vest title. 35 U. S. C. §202(a) (emphasis added). Stanford reaches the opposite conclusion, but only because it reads “retain” to mean “acquire” and “receive.” That is certainly not the common meaning of “retain.” “[R]etain” means “to hold or continue to hold in possession or use.” Webster’s Third . . .

The university folks advocating for Bayh-Dole not to be used as the law is written have a long history of grossly distorting the meaning of the statute. Here, the Supreme Court points out that Stanford’s version of “retain” is just the opposite of the meaning of the word–and getting to the opposite of the plain meaning qualifies as “gross distortion.”

Reimers cites one bit of what march-in is about–nonuse–and then makes it appear that the definition of “practical application” doesn’t mean what it does mean. According to Reimers’s argument, we are to read “practical application” to mean “use” only. The march-in at 203(a)(1) then would read: if a contractor or assignee is not “making an effort,” and not likely to make an effort, then march-in. But that reading throws out most of the definition of practical application. I know–that’s just like something Stanford would do, but it’s not the law. It’s a gross distortion of the law. I’d have a whole lot more respect for Niels Reimers if he admitted he is wrong about the law, saw that it wasn’t being enforced to protect the public, and that the public in the Xtandi case ought to be protected–but maybe march-in isn’t a good way to do that. I’d agree with that position.

March-in is third in line for Bayh-Dole tools to address unreasonable price. The first is the government license to practice (make, use, and sell) and have practiced. For public health matters, the government’s license allows it to authorize the production and sale of products based on any subject invention without all the fussies of worrying over 203(a)(1) citations to 201(f) definitions that Stanford-affiliated folks can’t bring themselves to read. The government license is unconditional, irrevocable, and does not require any procedures or notices to the contractor or delays while the contractor tries to explain how the government’s license was “intended” to be less than what it is, or that if the government actually relied on the rights it has, that would ruin the universities’ efforts to license preferentially to those companies all too eager to screw over the American public. Why, yes, perhaps the federal government relying on its license would do that, and yes, indeed, that might be a good thing, a good public policy goal, something that Congress might have intended in passing Bayh-Dole, and would approve now.

The second provision in line in Bayh-Dole to address pricing is the restriction on nonprofit assignment of inventions, 35 USC 202(c)(7)(A). If a nonprofit assigns a subject invention, then the assignee must comply with the nonprofit’s patent rights clause. The nonprofit’s patent rights clause requires that all income earned with respect to a subject invention must be used for scientific research or education, deducting only those amounts incidental to the administration of subject inventions, including sharing royalties with inventors. Assignments of inventions may be done in at least two ways. The first is an express assignment. There’s a document labeled “Assignment” and the owner of the invention signs under a statement of the form “I hereby assign all right, title, and interest in and to [such and such invention].” The second is an implicit assignment. There’s a document labeled “Exclusive License Agreement” and the owner of the invention signs a statement of the form “University hereby grants, and licensee accepts, an exclusive license to make, use, and sell Licensed Subject Matter, with the rights to sublicense and to enforce Licensed Patents, for the term of the last to expire of Licensed Patents.”

Courts have repeatedly held that exclusive licenses that convey all substantial rights in an invention (make, use, and sell) effect an assignment of the invention, even if title to the patents involved is withheld by the licensor. And it makes sense that such a finding holds for Bayh-Dole. If Bayh-Dole’s restriction on nonprofit assignments was restricted to just assignments labeled “Assignment,” then it would be simple to work around the restriction, using instead an implicit assignment in the form of an exclusive license to all substantial rights. The law then would have to do with what choice of labels nonprofits placed on written instruments that conveyed ownership of subject inventions. The idea would be that Congress intended for nonprofits to restrict the use of income earned with respect to a subject invention only when ownership was conveyed by an express assignment and not any other way of conveying ownership. That would be a pretty grossly distorted reading of Bayh-Dole. Hey–I know some folks who would be up for that reading. I’ll make you guess.

Universities routinely assign subject inventions using an implicit assignment that takes the form of an exclusive license for all substantial rights. That’s what UCLA did in the case of the series of compounds from which Xtandi was developed as a product. All substantial rights to Medivation. Under 35 USC 202(c)(7)(A), Medivation is subject to the requirement that all income earned with respect to the subject invention go to scientific research or education. That hasn’t happened. How does that matter for reasonable pricing?

If a for-profit company complied with the 202(c)(7)(A) requirement, we might think highly of them–making the effort to commercialize a new drug so that all the income would go back into more research (or education). If their new drug had an unaffordable price but that *all* their income was going back into research, well, maybe that would be “reasonable.” Not really, but maybe there’s something to it. More likely, however, profit-seeking companies would decline to get involved on that basis. They would want a non-exclusive license–they would choose to have competition (if not collaborators to help with development before competing). The competition would address pricing. That’s how Bayh-Dole works. That’s the Bayh-Dole framework. That’s the meaning embodied in the 202(c)(7)(A) requirement.

When the federal government uses its license to practice and have practiced, it can acquire a drug–Xtandi, say–for all government purposes. Public health is a government purpose (as evidenced, in one instance, by the billions that the federal government pours into public health research). If a company with an exclusive license (=assignment) offers product at a price comparable to what the federal government would pay to authorize a company to make that product for government purposes, then maybe the government would not need to use its license, because it would obtain a competitive price even without actual competition. If the company prices higher than it would with competition, then the government uses its license, introduces competition, and the price comes down for everyone, or the company sells way less product.

Similarly, if a company with an exclusive license (=assignment) sells at a high price and dedicates all income to scientific research or education, then although the price may be “high” since the money is going to an overtly public purpose, then perhaps the price is “reasonable” and everyone just gets on with it. The company is not enriched and passes the money on for use in research and education. The drug in question may not be affordable, but at least the terms (including price) are reasonable because Bayh-Dole all but insists that such a use of income is reasonable–even required. If the company declines such an exclusive license (=assignment), then we are back to multiple licensees who can do whatever they want with their income earned with respect to the subject invention rights they have licensed (non-exclusively), and once again we have competition and prices adjust accordingly. No price controls–instead, there’s free competition incentives, and free competition addresses reasonable pricing–pricing that happens when patents are used to promote free competition and utilization rather than to promote a sole source and suppress utilization but for whatever the sole source decides to use, if anything.

We are nearly done. I’m proud of you for getting this far. Scams are, by their nature, covert and built on deception and distraction. It takes time to see them for what they are. There’s only a bit more to finish with Reimers’s second op/ed:

This petition grossly distorts the Bayh-Dole Act. If it were adopted, innovators would know that anyone could ask the government to “march in” to allow copiers to undercut them in the marketplace.

We have dealt with Reimers’s claim that the Xtandi march-in petition distorts Bayh-Dole. It doesn’t. Reimers distorts Bayh-Dole, even grossly. “Innovators” know that the government can march-in right now–it’s not some future subjunctive. But more gross distortion coming here. Bayh-Dole requires a finding of unreasonable terms before any “copiers” could “undercut” the price on offer. Reimers is in effect arguing that “price-gougers would know that companies willing to offer similar product on reasonable terms could be authorized to do so.” What Reimers doesn’t point out (maybe he doesn’t know) is that in march-in, the contractor forced to grant licenses can expect to *get paid*. March-in licenses are to be on “terms reasonable under the circumstances.” Here, as with the definition of practical application and its “reasonable terms,” we would expect “reasonable” terms to involve royalties indexed to the amount of making, using, and selling the new licensees do. The “innovator”–that first company with the unreasonable pricing problem–gets paid to allow competition, and so has an advantage over those competitors, as they have to pay a reasonable surcharge on their activity. The “innovator” then has money to spare and therefore could lower prices based on what the competition pays to get a share of the action. These new competitors then know that the “innovator”. might undercut their pricing in the “marketplace.” Overall, however, the price differences likely would not be so much that other factors may still come into play–availability, form factor, method of delivery, combination with other drugs, brand, post-sale support.

March-in is not merely about “copiers” “undercutting” reasonable pricing. It’s about what should happen when the innovator’s pricing is not reasonable and competition would lead to benefits being available on reasonable terms.

Given that with the government license, “innovators” don’t get a darned thing–the government license is royalty-free–and with the nonprofit patent clause restriction on use of income, an “innovator” doesn’t get to keep what it has been paid–one would think that of all the incentives for free competition in Bayh-Dole, companies should desire march-in the most. Make the government find the competition, settle on reasonable terms with those competitors, and make them pay the “innovator.” The “innovator” could even stop producing its product and just take a cut of all the sales activity in the market without having to spend a dime itself, other than on the next big product. If one were looking at profit not revenue, having the government set up your market-wide non-exclusive licensing program for you isn’t necessarily the end of the world.

Now another flight of dystopian fantasy:

No one would make investments under such conditions. Once again, taxpayer-funded discoveries — like the one that led to the prostate cancer drug the petition concerns — would languish in labs.

What? Companies are so scared of competition that they stop competing whenever there’s prospect of competition, even price competition? No, it can’t be. Apple doesn’t compete on price. It introduces products that it has invested billions in that “copiers” can “undercut” in the “marketplace.” Apple even introduces products that apparently use Caltech inventions made in work receiving federal funding, without exclusive rights, and without even a Caltech license–neither of which appears to have affected Apple’s decision about what to use. Rather than claiming success of its tech transfer efforts, Caltech goes off and sues Apple for infringement. Hardly a use of the patent system to promote utilization. One might think, even, that Caltech operates outside of the patent property right Bayh-Dole provides for contractors acquiring subject inventions. Another gross distortion of Bayh-Dole.

If Reimers means by “such conditions” the conditions he described–an innovator getting undercut (on price, say) by copiers–then perhaps in the abstract that may be true for some innovators. What “innovators” would those be? Perhaps the “innovators” that will invest in something new that has been discovered in work claiming a public benefit may result only if they can expect to price gouge the public or suppress use. So, speculators on public health, patent trolls aiming to extract rents from everyone involved in public health. Investors building pyramid schemes to attract more investors hoping to exploit the value of the patent to channel demand into a pay garden of unlimited delights. We might think then that Reimers argues that Congress intended that the response to a perceived loss of U.S. leadership in global technology was to entice more price-gougers, speculators, trolls, and pyramid schemers into the development of research funding at least in part by the federal government.

In a not at all distorting reading of Bayh-Dole, we might posit that Congress intended the public protections that gave incentives for free competition to limit the participation of price-gougers, speculators, trolls, and pyramid schemers. If these are the only companies that, say, UCLA or Stanford can find willing to pay big-time for university exclusive licenses (=assignments), and Reimers argues that protecting the public will preclude universities from licensing to just these folks and so the public cannot be so protected, then it is Bayh-Dole that’s a policy failure, and we need to start all over.

The Biden administration can prevent a catastrophic drop in private-sector research and development investment by rejecting this latest march-in petition.

The Biden administration should use the government license in Bayh-Dole immediately. It can save hundreds of millions of dollars a year on Xtandi alone, taking money wrongfully gained by a Japanese company. Repeat for the handful of drugs that also have a Bayh-Dole patent behind them, and introduce competition–and technology standards–into the pharmaceutical industry. There’s nothing to indicate that private-sector investment will drop “catastrophically” because there’s more competition in matters of public health. Less than 10% of drugs on the market have a patent connection to federal funding. That means 90% of private-sector investment in new drugs does not depend on Bayh-Dole patenting. (Stevens et al. give a summary of their findings–looking for everything government-related that they could.) Other studies argue that the vast majority of drugs on the marketplace have been developed with research obtained from federally supported work. For instance, a study looked at the 210 drugs approved between 2010 and 2016 and found all of them connected to published research funded by the NIH. Most of that research has not been patented. The idea that comes to mind is that technology transfer works just fine in health-related fields without every university patenting everything it can and each one holding out, with each patent, for a single exclusive licensee (=assignee).

If we refine Reimers’s argument for him, it is this: if universities are made to comply with Bayh-Dole, then they will not be able to attract the price-gougers, speculators, patent trolls, and pyramid schemers that they have feasted with for the past forty years. Investment by these folks in health products might indeed suffer a “catastrophic drop.” That, perhaps, would be a good thing. 90% of the drugs developed in the Bayh-Dole era have had nothing to do with Bayh-Dole patents, even if they have made use (free use) of other results of federally supported research.

Of the less than 10% of drugs that do have patents that cite federal funding, the problem is that a number of them involve non-compliance with Bayh-Dole, price-gouging, restricted supply, or patent trolling. In some high visibility cases, such as the University of Pennsylvania patents covering a trick with mRNA to limit adverse reactions by the human immune system, no one needed an exclusive license. Penn declined to license to the inventors’ startup and licensed exclusively (=assigned) instead to a small biotech company in Wisconsin, which then through a series of restructurings and new startups, eventually sublicensed to Moderna (a Harvard-affiliated startup) and to BioNTech (where one of the Penn inventors ended up as a VP). The “innovator” company didn’t develop any vaccine itself. It just distributed the rights non-exclusively and pocketed millions–what Penn could have done itself, and what Reimers, the old technology transfer expert, claims would never, ever happen.


This entry was posted in Bayh-Dole and tagged , , , . Bookmark the permalink.