No, the government cannot seize, break or ‘bypass’ pharmaceutical patents — even for COVID-19

A seemingly endless stream of politicians is calling for the government to seize or bypass pharmaceutical patents. Thirty-four state attorneys general demanded that the government exercise march-in rights on remdesivir, a medicine used in treating COVID-19, while Sen. Elizabeth Warren (D-Mass.) called on the incoming Biden administration to use “compulsory licensing authority” to “bypass patents for pressing public health needs” on other medicines. These proposals and others fatally misconstrue the law.

There are three legal avenues at play. The first two apply only if federal funding is involved under the Bayh-Dole Act of 1980: march-in rights and a nonexclusive government license. The third is Section 1498 of Title 28 of the U.S. Code, governing the judiciary, that gives patent owners a remedy in the court of federal claims if the government infringes their property rights.

Arguments for how Bayh-Dole applies to remdesivir are unconvincing, especially when the U.S. Army as funding agency appears to be disclaiming rights. While there might be evidence that government scientists participated in the invention, an Army attorney flatly asserted that this “did not qualify USAMRIID as a joint inventor of the compound.” The state AGs assert that Gilead “received substantial federal funding” to develop remdesivir, but as Joe Allen and Professor Adam Mossoff point out, this was after Gilead patented the drug. 

Oddly everyone misses how Bayh-Dole might apply: if remdesivir was first actually reduced to practice under Army funding. Under U.S. patent law an invention may be patented even if it was only constructively reduced to practice — described in the patent application such that a “person having ordinary skill in the art” could practice it without undue experimentation. But Bayh-Dole covers any inventions “conceived or first actually reduced to practice in the performance of work under a funding agreement . . . .” Courts have confirmed that this includes issued patents whose invention was not actually made in the real world until later under federal funding.

Even if Bayh-Dole applies to remdesivir, there is no legal pathway for the government to seize or bypass Gilead’s patent.

March-in rights under Bayh-Dole’s Section 203 only authorize the government to grant new licenses if the original funding recipient fails to take steps to bring the invention to the market (achieve “practical application”) or reasonably satisfy health or safety needs. (Two other triggers are not relevant here.) March-in rights do not authorize the government to “seize” or “break” the patent. The contractor still owns the patent – and can enforce it against other parties – and the existing licenses remain in effect. In many cases, a march-in license will not result in more drugs getting to market faster anyway because the new licensee will have to scale up under strict drug manufacturing and distribution regulations. By the time it does so, the original manufacturer will likely have solved whatever problems it had. 

Much mischief resulted, however, from a 2001 law review article – and accompanying Washington Post op-ed – by two academics who had no apparent expertise in tech transfer law. Peter Arno and Michael Davis believed they had stumbled on a shocking fact: few had noticed that an “obscure” law called Bayh-Dole put in place retail price controls for government funded inventions. 

Their argument turned on the statutory definition of “practical application”: “[to practice the invention] under such conditions as to establish that the invention is being utilized and that its benefits are . . . available to the public on reasonable terms.” (emphasis added). Understandably, this may sound like retail prices of a product covered by the patent must be “reasonable.” But tech transfer practitioners knew this meant the license terms a federally funded patent owner offered to manufacturers to bring the product to market. The contractor should not demand such onerous terms that its licensing offer is a sham.

Bayh-Dole is not an obscure law, nor does it authorize retail price control. Arno and Davis’ “insight” was based on a further mistake: they cited congressional debate about an earlier tech transfer bill that was replaced in its entirety by the bill enacted as Bayh-Dole. Bob Dole corrected the record as part of his introductory remarks to an NIH public march-in rights petition hearing. Egregiously, the state attorneys general base their remdesivir march-in rights argument on this discredited Arno & Davis article.

The government license under Bayh-Dole’s Section 202(c)(4) will only benefit the general public when the government is ready to take the precipitous step of providing drugs as a government service. While many support this, others will argue that this is a slippery slope to “socialized medicine.” Like march-in rights, this is not a seizure or breaking of the patent. It is simply a license the contractor had to grant as a condition of the government funding agreement.

The final pathway, Section 1498, is often mischaracterized as a compulsory license. It is in some ways the complete opposite. Due to sovereign immunity, patent owners could not sue the government for infringement into the early 20th century. However, Congress can enact statutory exceptions to this general principle. Section 1498 does just that by giving patentees an express right to sue the government for full and fair compensation in the Court of Federal Claims. Thus to call this a compulsory license turns the statute on its head. The government is no more “authorized” to infringe patents than are private individuals.

The siren song of free or low-cost pharmaceuticals continues to lure academics and policymakers to ignore the carefully crafted balance of incentives in our legal system. Ultimately, drug prices are driven more by other factors in the health care system than by patent rights that incentivize lifesaving innovations. The incoming Biden administration would do well to listen to experienced voices in drug development and take time to understand the laws as they actually stand right now, rather than follow these mistaken calls for seizing, breaking or bypassing drug patents.

Sean M. O’Connor is professor of law and executive director at the Center for the Protection of Intellectual Property (CPIP) at George Mason University, Antonin Scalia Law School.