Xtandi demonstrates how Bayh-Dole is a do WTF you want law, 2

The first part of this article is here.

To explore these issues, let’s take a look at Xtandi, a prostate cancer drug that has received press recently in the debate over high drug prices. Xtandi sells in the U.S. for $88 a pill, and a patient takes 4 pills a day. Things add up–over $120,000 a year, per patient. No wonder Xtandi has racked up over $2B annually in sales. But in Canada, Xtandi sells for just $12 a pill. And a Canadian generics drug manufacturer, Biolyse, says it could make the drug for $3 a pill–a cost to a patient (or patient’s insurer) of $4,000 a year. So the U.S. price is 8x the Canadian price, and 30x what the drug would sell for with competition.

If one is a shareholder in Medivation (or, now, Pfizer, which acquired Medivation through its subsidiary Montreal for $14B), then these are happy times. If one is a shareholder in a medical insurer, then things are dire as usual. If the federal government is willing to pay the higher price, then taxes go out to the corporation that has fronted the clinical development of the drug. As far as the taxes go, then, there is a civic discussion regarding how we spend our collective money. At present, the decision is that it’s really great to see shareholders getting such a return on the risk they took buying shares. As a result of their risk-taking, a beneficial drug is on the market and people’s lives are being helped. It’s all good.

Similarly, if one is a patent owner and has licensed the patent to Medivation for a royalty on sales, equity, and other milestone payments and reimbursements. And that’s what UCLA is, since enzalutamide–the compound Xtandi is based upon–was identified for clinical use at UCLA, with federal funding. So the UCLA invention is a subject invention under Bayh-Dole. There is plenty of complexity behind the scenes to get at what has gone on, and I have tracked down only some of it. But it’s worth taking a look to see how things have developed, and how Bayh-Dole has been used, or, as it were, exploited or ignored.

UCLA took ownership of patent rights (in the name of The Regents of the University of California–the corporation that operates the UC campuses) to the use of a class of compounds that appeared to have anti-cancer properties in 2003. One of the scientist inventors went to the founder of Medivation, hoping to interest him in licensing the invention. Medivation had been set up to hunt for promising compounds to develop into therapeutic drugs. Medivation already had one drug for Alzheimers in early clinical trials (Dimebon, which failed Stage 3 trials in 2010). UCLA, acting for The Regents, then licensed the patent rights in 2005 exclusively to Medivation, a company that had been set up to hunt for profitable drugs to develop as commercial products and already had one candidate drug under management.

It is worth pausing here. The “invention” consisted of a class of compounds, called the “RD Series,” and the patent as well as claimed the use of those compounds in pharmaceuticals, and the application of this class of compounds to cancer treatment in mammals, and of course also expressly in humans, and claimed using a variety of mechanisms to do so. The “invention” therefore was no one thing–it is a whole broad territory of things. According to Medivation, there were about 170 compounds involved in the RD Series. Over a period of years, Medivation funded further research at UCLA that added to the compounds that Medivation licensed.

The compound Medivation selected to develop was 162 prime, which they called MDV3100. That means there are 169 or so other compounds that aren’t being developed, that no one else can develop for prostate cancer or any other cancer or any other condition in mammals. UCLA’s exclusive license to the patent on the “invention” then may enable the development of one compound, but it also authorizes the nonuse of 169 others. Under Bayh-Dole, this kind of licensing is apparently outside UCLA’s patent property rights under 35 USC 200, fails the definition of practical application at 35 USC 201(f), and ought to spark a march-in by the federal agency under 35 USC 203 for all those other compounds–except federal agencies don’t have to march-in, and so don’t.

But there’s more. The same UCLA scientists that developed the RD Series also developed a second class of similar compounds, called the “A Series.” They didn’t tell Medivation about the A Series compounds (A51 and A52)–not during the licensing talks, not during any patenting work, not in the research Medivation sponsored. One of those compounds, A52, is apparently chemically almost the same as RD Series compound 37. UCLA licensed the A Series compounds to Aragon Pharmaceuticals, a company formed by two of the UCLA inventor-scientists to create a prostate cancer treatment–one that might compete with Xtandi.

When Medivation discovered UCLA’s patent applications for the A-Series compounds and that UCLA had licensed them exclusively to Aragon, Medivation sued UCLA. That was in 2009. In 2012, a court ruled that UCLA was within its rights, had narrowly defined what it had licensed and what it was obligated to report to Medivation under its research agreements. The crux of it came down to a research contract with an effective start date of November 1, but with interactions starting in August, and the work to synthesize an A Series compound taking place on October 28. Since UCLA committed only to reporting inventions “conceived and reduced to practice” in the research with Medivation, clearly the A Series had been “conceived” before the effective date of the sponsored research agreement. Medivation loses. Sneaky? Perhaps.

Aragon went on to sell out to Johnson & Johnson for up to $1B in 2013. Compound AR52 is in clinical trials. All J&J wanted was the AR52 rights (also called AR509). What had Aragon done to deserve $1B? Here’s what J&J reported in its 10-Q:

During the fiscal third quarter of 2013, the Company completed the acquisitions of Aragon Pharmaceuticals, Inc., a privately-held, pharmaceutical discovery and development company focused on drugs to treat hormonally-driven cancers. Under the terms of the agreement, the Company made an upfront cash payment of $650 million. Additional contingent payments of up to $350 million may be paid in the future based on reaching predetermined milestones.  The purchase price for the acquisition including the fair value of the contingent consideration of $0.2 billion was allocated primarily to purchased IPR&D for $0.8 billion and goodwill for $0.2 billion.

$800M for IP rights and development and $200M for “goodwill”–i.e., recognition of the brand and capabilities of the company. That’s a heck of a lot of “goodwill” for a four-year old company with an exclusive license to a compound–other than that the company also includes two of the key co-inventors (as it were) of the AR52 compound. The co-inventors didn’t get that from UCLA when UCLA made them assign their rights to the university.

We might pause here as well to note some interesting elements of the story thus far. Despite the claims that a company won’t invest without an exclusive license, clearly UCLA managed to induce Medivation to take an exclusive license to some but not all of the compounds that showed promise for fighting prostate cancer–and using pretty much the same strategy. And Aragon took a license knowing that other compounds were being developed that would compete with AR52. Sure, Aragon was the inventors’ company, but that did not prevent it from getting investors and getting bought out by a major company.

If UCLA had defined the “invention” more broadly to include all the compounds that worked roughly the same way, then Medivation would have negotiated to obtain an exclusive license for all such compounds–A Series and RD Series. This may well have been an even better position for Medivation. But why should either company hold exclusive rights to the compounds that they are not developing? Why should those compounds be unavailable for development? You can see, perhaps, the limitation in talking about an “invention” or assuming that a patent right covers some singular thing. This is a basic inadequacy in the language used in patent policies and public discussions. If we want “inventions” to be used, then we will also have to pay attention to the structure of those inventions, their material manifestations, and how these inventions are represented in patent claims. In the case of Xtandi, one compound gets developed as a product, and many others suffer nonuse. And only one company–now Pfizer–has the right to explore the use of that compound–or any of the others in the RD Series–for any other cancer uses. Certainly, a benefit of the “invention” is available. But just as apparently, many benefits of the invention are not available as a consequence of the exclusive license.


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