On the last day of scheduled oral arguments in the term, the Affordable Care Act — the always controversial “Obamacare” statute — returns yet again to the Supreme Court, although many of the justices and their clerks will probably be wishing it weren’t so.
The consolidated patent cases Sandoz v. Amgen and Amgen v. Sandoz involve the complex interaction between the Patent Act and the Biologics Price Competition and Innovation Act, which is an approximately 17-page subchapter contained in the larger 906-page ACA. That larger statute has not been hailed for precise drafting or ease of application. This new litigation will certainly not improve the reputation of the statutory behemoth.
The cases arise out of (i) Sandoz’s petition for certiorari on two questions and (ii) a quite rare “conditional cross-petition for certiorari” filed by Amgen that presents a third question. In January, the court granted both the petition and the conditional cross-petition; consolidated the cases; set a special briefing schedule allowing longer briefs in less time than normal; and scheduled the case for the last day of arguments in the term.
This initial description only hints at the truly startling complexity of these consolidated cases. Together they are a giant Russian nesting doll, a matryoshka, of complication. Even though I have been teaching and writing about patent law for years, I find these consolidated cases daunting. I can only imagine what the poor law clerks assigned to work on them are thinking.
Let’s take apart the nesting doll. I’ll describe five layers of complexity and apologize in advance for not doing justice to any particular layer.
First, at the highest level of generality, these consolidated cases are another battle in a long-running policy war over the optimal level of patent protection for medicines, with the battle lines drawn as usual. The generic-drug company, Sandoz, seeks slightly less patent protection, which would lower drug costs and provide short-run benefits to consumers and health insurance companies. The research pharmaceutical company, Amgen, wants a bit more protection, which is likely to provide better medicines to treat cancer and other maladies, but only over the long run.
The best that can be hoped for is that the justices understand that the case is not one-sided at the policy level. The amicus briefs can help in this regard, but only if amici on both sides are taken seriously. An amicus brief by America’s Health Insurance Plans supports a bit less patent protection as a way to “control spiraling health care costs.” A professors’ amicus brief written by noted patent attorney Kevin Noonan (who is Justice Sonia Sotomayor’s ex-husband) reminds readers that “[i]nnovation in medicine over the last century is responsible for profound improvements in public health,” and that continued innovation in medicine “depends on the availability of meaningful patent protection.” Which brief is right? Both, of course; they are two sides of the same story. Patent protection increases costs today to improve lives tomorrow. It’s always been a controversial and difficult tradeoff.
Second, this case also involves complex technology – which is why the litigation involves not merely the general Patent Act but the interaction between that statute and the provisions in the ACA designed to regulate “biologics.”
Biologics are a class of bio-engineered drugs that are far more complex and hard to produce than the typical “small molecule” drugs patented in the 20th century. For example, just 50 atoms are contained in a molecule of loratadine, also known by the trade name “Claritin” (chemical formula: C22H23ClN2O2), which was patented in the 1980s and became a popular “blockbuster” drug in the 1990s. By contrast, the drug at issue in this litigation — filgrastim — contains several thousand atoms in each molecule (formula: C845H1343N223O243S9).
Typically (or at least typically in the 20th century), when the patent on a drug expires, generic companies can enter the market with exact copies of the previously patented drug. That happened in the 2002 for loratadine, and the drug’s price fell dramatically as ever more competitors entered the market. The sheer complexity of biologics makes them very difficult for generic drug companies to reproduce exactly. Often the best that can be done is to produce a “biosimilar,” which, as its name suggests, is not identical to the original biologic drug. The imprecision, however, creates a regulatory problem because the Food and Drug Administration must approve any new drug before it is sold.
The biologics provisions in the ACA were intended to streamline the FDA approval process for biosimilars and also to provide a clear and structured process for addressing the patent issues that were almost certain to arise as competitors tried to copy these drugs. Now, the very first time the provisions have been used, a three-judge panel for the U.S. Court of Appeals for the Federal Circuit split three ways, and this complex piece of Supreme Court litigation followed. So much for clarity and simplicity!
Third, having dealt with the basic policy choices and the underlying technology, we can now turn to the legal issues in the case. The new biologics statute includes two key structures. Subsection (k) (of 42 U.S.C. § 262) creates a process by which the FDA can license biosimilars for commercial marketing. Subsection (l) provides a process by which the applicant for a biosimilar license (i.e., the generic pharmaceutical company) can exchange information about patents with the company that first obtained FDA approval to market the original biologic drug (i.e., the research pharmaceutical company).
One of the very first requirements in that statutory “exchange of information” process is that the applicant for a biosimilar license – here Sandoz – “shall provide to the [research pharmaceutical company] a copy of the application submitted [to the agency] under subsection (k).” The word “shall” in the statute might suggest that Sandoz is required to provide a copy of its biosimilar application to Amgen, but the divided panel of the Federal Circuit ruled disclosure of the application optional based on two other statutory provisions that prescribe consequences when the “applicant fails to provide the application” to the research pharmaceutical company. Those two sections, the Federal Circuit reasoned, would be rendered “superfluous” if disclosure were mandatory.
The Federal Circuit thus inferred from Congress’ expression of some specific consequences for failure to disclose a biosimilar application an intent to exclude other additional remedies, such as injunctive relief compelling compliance with the requirement that the applicant “shall provide” the application. Such an inference parallels the reasoning underlying the controversial “expressio unius” canon of statutory construction – that the expression of one thing implies the exclusion of others.
That logic has been described as “one of the most fatuously simple of logical fallacies.” If I tell my students that I’ll fail anyone who cheats, I do not mean to foreclose the possibility of other, additional consequences, such as expulsion from the university. Perhaps the Federal Circuit’s inference is more plausible as applied to this particular statute, but there’s a lot of force to Amgen’s argument that the word “shall” bespeaks a mandatory, not optional, duty.
Fourth, the statute also requires that the biosimilar applicant (here Sandoz) “shall provide notice to [the research pharmaceutical company, here Amgen] not later than 180 days before the date of the first commercial marketing of the biological product licensed under subsection (k).” Sandoz sent the notice while its biosimilar application was still pending before the FDA so that, Sandoz hoped, the biosimilar could marketed as soon as the FDA licensed it.
The Federal Circuit, however, concluded that Sandoz had sent the notice too early. Relying on the structure of the statute, the Federal Circuit thought that the notice should logically have “to follow licensure, at which time the product, its therapeutic uses, and its manufacturing processes are fixed.” Thus, the court held that “notice, to be effective under this statute, must be given only after the product is licensed by the FDA.” That holding meant that Sandoz had to wait to introduce its licensed product until 180 days after the FDA’s approval, a delay to which Sandoz objected.
Here too, the Federal Circuit’s logic is open to question, though the issue is difficult due to poor statutory drafting. The Federal Circuit might be right in thinking it sensible to require the biosimilar applicant to send notice after the FDA licensure; the applicant could then inform the other company of the precise conditions of the FDA’s license. The statute, however, does not require that the applicant provide notice about the terms and conditions of the license or about the manufacturing processes approved by the FDA. Rather, the statute – quite incredibly – merely requires “notice” but nowhere expressly specifies what information should be included in the notice.
The title of the statutory sub-paragraph, “Notice of commercial marketing,” together with the requirement that the notice must precede the marketing by 180 days, suggests that the notice merely has to include (as Judge Raymond Chen argued in dissent below) a statement of the “intent to begin commercially marketing the biosimilar product.” And because such a notice merely recites intended actions expected to take place in the future, the statutory language could reasonably be interpreted as permitting the notice to be contingent on future events, such as expected issuance of an FDA license. That interpretation is also consistent with the structure of the sentence because, as Sandoz argues, the entire phrase “the date of the first commercial marketing of the biological product licensed under subsection (k)” appears only to define “a specific point in time” in the future at which time the product will be “licensed.”
Fifth, the last of the questions presented in the case is whether a court can enforce the 180-day notice requirement – however that is calculated – by enjoining the marketing of the drug until notice is given. Simply and generally stated, Sandoz is arguing that a line of Supreme Court cases restricting implied private rights of action should also restrict the equitable powers of federal courts. This issue, unlike the others in the case, could have an enormous impact on many other areas of federal law, including federal administrative law.
On this hugely important issue, the parties spend a significant amount of time debating a position articulated in a Federal Circuit biosimilars case decided about a year after the decision below in this litigation. (If you are wondering why the parties in this case are debating judicial reasoning expressed in a different case, the answer is that virtually nothing is easy or straightforward in this litigation.)
Writing for the court in that other case, Amgen Inc. v. Apotex Inc., Judge Richard Taranto relied on Supreme Court precedent to rule that the federal courts’ “equitable jurisdiction is not to be denied or limited in the absence of a clear and valid legislative command.” As Justice Neil Gorsuch’s former colleague on the U.S. Court of Appeals for the 10th Circuit, Judge Michael McConnell, explained in Simmat v. United States Bureau of Prisons, an express or implied private right of action is unnecessary for injunctive relief because “[e]quity … provides the basis for relief — the cause of action, so to speak — in appropriate cases within the court’s jurisdiction.”
As support for the view that equitable relief needs no additional statutory authorization, Judge McConnell cited one of my articles. Although that article discussed non-statutory equitable relief against government officers (an important issue in administrative law), such relief was traditionally available only because officers acting beyond their lawful authority were viewed as being non-governmental (and thus not protected by sovereign immunity). My article shows not only that I have academic views on this issue (views that readers are free to discount), but also that the general issue has implications far outside the bounds of patent law. That alone should make the court cautious about contravening long-standing tradition concerning federal court power to enjoin violations of federal law in the course of deciding this complex patent case.
* * *
In closing, I’ll offer one final point just to prove how complex the case is: Every question presented by the parties is moot with respect to this case!
Take, for example, the dispute over whether the ACA requires Sandoz to turn over its biosimilar application to Amgen. Is Amgen still waiting to get that application? No! Amgen sued Sandoz for patent infringement and got the application in discovery. Or consider the issues concerning when Sandoz should have given notice to Amgen and whether the courts can enjoin Sandoz for failing to give that notice. Those too are solidly in the rear-view mirror because the injunctive relief ordered below has now expired.
So why is this case still being litigated? Sandoz argues that the case is “capable of repetition yet evading review” – a supposedly narrow exception to mootness deployed most famously in Roe v. Wade.
Perhaps that exception can cover part of this case (though the parties’ briefing on mootness leaves much to be desired), but let’s hope that the justices are not too eager to decide every question possible. For example, if the Supreme Court rules that the Federal Circuit miscalculated the 180-day notice period, the fifth issue may well be moot, because Sandoz would not again be subject to the same action (i.e., an injunction holding its product off the market for 180 days after the issuance of a biosimilar’s FDA license).
As this case is about medicines and health care, the court would do well to remember the maxim often attributed to the Hippocratic Oath: “First do no harm.” That’s good advice, because the chances for unintended mischief here are great. It’s a hugely complex case involving the very first application of a new statute; it’s being argued on the last day of the last argument session of the term; it presents issues that most of the justices and their law clerks may not have considered often in the past; and finally it involves provisions of the notoriously complex and imperfectly drafted ACA. I expect, and hope, that the oral argument will show the justices being cautious lest they do harm.