or nearly a century, judges and lawyers debating the First Amendment’s freedom have relied on a familiar metaphor: the “marketplace of ideas.” The image comes from Justice Oliver Wendell Holmes’s dissenting opinion in Abrams v. United States: Rejecting the Court’s affirmance of wartime convictions under the Sedition Act, Holmes urged that “the best test of truth is the power of the thought to get itself accepted in the competition of the market.”
In some free speech cases, the “marketplace” is not merely a metaphor—it is the actual fact of the case. In regulating interstate commerce, Federal agencies sometimes venture into the constitutionally dangerous territory of regulating speech. This is doubly dangerous: By restraining speech, the government not only undermines the nation’s commitment to the First Amendment, but it also prevents important facts from reaching the market, which in turn forces Americans to bear some risk in buying goods and services without the benefit of all relevant information. Put another way, a free market depends on the widest possible dissemination of relevant commercial information, the suppression of which leads inevitably to market failure and, eventually, government control. As Louis Brandeis, the lodestar of progressive reformers, stressed a century ago, “sunlight”—not suppression—is “the best of disinfectants.”1 These are precisely the questions confronting the Federal courts as they hear cases arising from the Food and Drug Administration’s policy on discussion of “off-label” uses of FDA-approved drugs. Instead of ensuring that the general public is fully aware of the benefits, risks and dangers of drugs, the FDA’s policy, as currently enforced, threatens to block truthful information from reaching the market.
Off-label drug use is essential to public health. Prescribing off-label uses is entirely legal, even preferred in some circumstances. In 2006, the Archives of Internal Medicine published a study that found that 46 percent of all prescriptions for cardiac conditions were off-label.2 But under the FDA’s policy, simply providing information about off-label use, even when such use is the “standard of care”, can be illegal. Recognizing this paradox, a prominent Federal court recently curtailed the FDA’s attempt to enforce its “off-label” policy, in a case titled United States v. Caronia. The FDA’s policy may well be destined to reach the Supreme Court, either on appeal in Caronia or in another case. But for now, Caronia presents the most significant decision in this crucial debate. Let’s trace the path to this case.
n 2002 and 2005, the FDA approved the use of Xyrem to treat conditions related to narcolepsy. But Xyrem, like many drugs, carried with its benefits the possibility of serious side effects. Orphan Medical, the pharmaceutical company that developed Xyrem, hired sales consultants to promote Xyrem to physicians, and Alfred Caronia was its sales consultant for the New York counties of Queens, Nassau and Suffolk. Caronia, in turn, started a “speaker program” in which Orphan would pay physicians to promote Xyrem to other physicians.
Because Caronia’s and the employed physicians’ discussions with customers might eventually turn to not just Xyrem’s FDA-approved uses, but also its off-label uses, Orphan established procedures to handle doctors’ questions. If Orphan-employed physicians were asked about Xyrem’s off-label uses, the doctors could answer them on the spot, and they often did so by reference to their own experience prescribing the drug. But if Orphan’s non-doctor sales consultants, such as Caronia, were asked about its off-label uses, they would complete a “medical information request form” and send it to Orphan, which would in turn respond to the inquiry.
In October 2005, Caronia discussed Xyrem’s off-label uses with Dr. Stephen Charno. After mentioning its FDA-approved uses, Caronia further noted that Xyrem might also be effective against “insomnia, Fibromyalgia[,] periodic leg movement, restless leg [syndrome]”, and that researchers were also investigating its effectiveness against Parkinson’s Disease and Multiple Sclerosis. On other occasions, Caronia and an Orphan-employed physician, Dr. Peter Gleason, suggested that Xyrem could be used by patients younger than age 16, which was not an FDA-approved population for the drug.
Federal investigators promptly learned of those discussions, because Charno, the putative customer, was actually an FDA informant. Two years later, a Federal grand jury indicted Caronia on two counts: introducing a “misbranded” drug into interstate commerce, and conspiracy to introduce a “misbranded” drug into interstate commerce. Following a jury trial in October 2008, Caronia was found not guilty on the former count, but he was convicted on the conspiracy count. For this he was sentenced to one year of probation, 100 hours of community service, and charged a small fine.
“Misbranding” is the term at the center of Caronia’s trial, but more importantly it is the heart of the FDA’s policy on off-label uses. For while the FDA permits the sale or prescription of drugs for off-label uses, it prohibits the manufacturers from marketing drugs for such uses. The FDA’s organic statute, the Food, Drug, and Cosmetic Act, does not specifically ban pharmaceutical companies or doctors from discussing drugs’ off-label uses. But the Act does prohibit “the introduction or delivery for introduction into interstate commerce of any” drug that is “misbranded.” And the Act deems a drug “misbranded” if, among other things, “its labeling is false or misleading in any particular” or its labeling fails to bear “adequate directions for its use.”
Because a drug’s “uses” might include not just the FDA-approved use, but also its off-label uses, the FDA uses those provisions as the statutory “hook” to deter off-label use, even though the off-label use itself is legal, and in many cases may be the standard of care. In regulations interpreting and administering the Act, the FDA stresses that a company’s “intended uses”, the uses for which its labeling must include adequate warnings and information, can include the uses conveyed by the oral and written representations of the pharmaceutical company’s representatives.
hat, in a nutshell, is the chain of statutes and regulations that brought Caronia into the government’s crosshairs. It is not, on its face, necessarily a strained reading of the statute; the Food, Drug, and Cosmetic Act, like so many Federal statutes, contains imprecise terms that leave relevant agencies wide discretion to interpret statutory terms and fill statutory gaps. But the statutes and regulations tell only part of the story. Because whether or not the FDA’s interpretation of the statute is plausible, its aggressive enforcement violates the Constitution.
Early in Caronia’s criminal trial, before the government and defendant began even to argue the facts before the jury, lawyers for Caronia and the Orphan-employed doctor, Dr. Gleason, moved to dismiss the charges. They argued that the FDA’s policy’s enforcement violated the First Amendment’s right of free speech. Dr. Gleason’s lawyers did not hesitate to highlight the unprecedented nature of the government’s charges against him:
This case concerns what appears to be the first prosecution in U.S. history of a physician for advocating the lawful use of a prescription drug to his professional colleagues. The treatments Dr. Gleason discussed were themselves legal, and the drug he allegedly promoted, Xyrem, had been approved by the [FDA].
To apply the FDA’s “misbranding” policies “to Dr. Gleason’s lectures to his colleagues violates the First Amendment’s command that ‘Congress shall make no law . . . abridging the freedom of speech.’” Those lectures, the lawyers continued, “are an essential part of the free flow of ideas protected by the First Amendment. Such discourse represents pure speech, which can only be regulated based on content in the rarest of circumstances.”
Dr. Gleason eventually pleaded guilty to a lesser charge. The pharmaceutical company itself also settled with the government, pleading guilty and agreeing to pay $20 million in penalties and victim compensation and to implement a “corporate integrity agreement.” But Caronia, the sales consultant, joined the doctor’s motion and pressed its other points himself, arguing that his statements were “protected under the First Amendment because he did not intend to defraud anyone” and that he sincerely believed the statements about Xyrem’s potential and believed in Dr. Gleason’s medical knowledge. More directly, he attacked the “convoluted logic” underlying the charges against him, and the FDA’s “misbranding” policy in general. The FDA, Caronia argued, believes that it
can regulate off-label speech if it is fraudulent speech; all speech that promotes off-label uses is fraudulent because it is not yet FDA approved; therefore, all off-label speech is criminal and not protected by the First Amendment.
The government, however, disputed Caronia’s characterization of the case. The FDA’s policy did not criminalize speech, it argued—the policy merely used speech as evidence to prove the pharmaceutical company’s intended off-label use of the drug:
[T]his is not a case in which Gleason is being held accountable for presenting purely scientific views on some bucolic campus, for prescribing Xyrem to his patients, or even solely for his speech. Rather, Gleason is charged with committing felonies with intent to defraud or mislead: causing Xyrem to be misbranded by promoting Xyrem for uses that were not adequately addressed in Xyrem’s labeling . . . and promoting Xyrem for uses which, absent appropriate warnings in the labeling, could be dangerous to the user’s health . . . and conspiring with Orphan and others to commit the foregoing violations.3
In short, because the government was using Caronia’s and Gleason’s “speech as evidence to prove a fraudulent conspiracy and misbranding”, the case is governed not by the First Amendment cases that Caronia invoked, but by cases governing the use of speech as evidence of other criminal acts. Noting that the Supreme Court had unanimously affirmed the validity of using speech for such evidentiary purposes, the government quoted another Federal court instruction that “[t]he question is whether this use of speech to infer intent, which in turn renders an otherwise permissible act unlawful, is constitutionally valid.”
Although the trial judge rejected the government’s characterization of the case and recognized the First Amendment issue at its heart, he refused to dismiss the government’s case against Caronia. Even if the FDA’s policy was subject to “heightened scrutiny” due to the free speech stakes, he reasoned, the policy was “narrowly tailored” to advance a “substantial” government interest, and therefore it passed constitutional muster. The case thus could proceed to trial before the jury, which convicted Caronia on the conspiracy-to-misbrand charge.
Caronia appealed to the U.S. Court of Appeals for the Second Circuit, where a three-judge panel reheard the constitutional arguments that Caronia and the government had advanced in the lower court. But where the trial court found that the balance of interests favored the government’s burdens on Caronia’s speech, the Second Circuit struck a different balance, and in so doing mapped precisely the right way for courts to parse these crucial constitutional issues.
First, and most importantly, in December 2012 the panel’s two-judge majority utterly rejected the government’s attempt to argue that it was prosecuting Caronia’s actions, not his speech. “Even assuming the government can offer evidence of a defendant’s off-label promotion to prove a drug’s intended use and, thus, mislabeling for that intended use”, the court wrote, “that is not what happened in this case.” As the record reflected, the government “repeatedly argued” that Caronia’s crime was his words, not his deeds; the government’s rationalization that it had seized upon Caronia’s statements merely to evidence Orphan’s intended off-label uses was “simply not true. . . . The government never suggested that Caronia engaged in any form of misbranding other than the promotion of the off-label use of the drug.” (The dissenting judge rejected her colleagues’ characterization of the case, agreeing that the government did in fact prosecute Caronia’s acts—his alleged conspiracy to sell Xyrem for misbranded uses—and that his speech was merely evidence of his intent to do so.)
Having identified the case’s true First Amendment stakes, the court’s majority then turned to the applicable constitutional tests for government regulation of this form of speech. On this point, the court relied on Sorrell v. IMS Health, a 2011 Supreme Court decision that defended the constitutional value of the commercial speech embodied in pharmaceutical marketing. Because the FDA’s prosecution of Caronia’s speech was content- and speaker-based (it centered on who he is and what he said), the Second Circuit concluded, the government’s position was subject to the constitutional standard known as “heightened scrutiny.”
And that was a degree of scrutiny that the government simply couldn’t survive: Applying the multi-factor test that governs Federal regulation of commercial speech, the court’s majority held that the FDA’s misbranding policy failed to sufficiently advance the broader goal of limiting the use of drugs for non-FDA-approved purposes. The policy barred pharmaceutical companies from touting off-label uses to doctors, but did not actually prohibit doctors themselves from prescribing the drugs for off-label uses. “The government’s construction of the [Food, Drug, and Cosmetic Act] essentially legalizes the outcome—off-label use—but prohibits the free flow of information that would inform that outcome.”
Similarly, in applying the other primary factor in the multi-factor constitutional test, the court held that Congress’s and the FDA’s policy was not “narrowly tailored” to achieve the government’s aim. Congress could have easily achieved its aim without heavily regulating speech: It could have simply banned off-label uses altogether, and left speech alone. “If the First Amendment means anything”, the court concluded, quoting the Supreme Court, “it means that regulating speech must be a last—not first—resort.”
he Second Circuit’s December 2012 decision in United States v. Caronia drew headlines, but it is by no means a novel case. Although the government already has announced that it will not appeal the Second Circuit’s decision to the Supreme Court, there is no shortage of other cases raising these First Amendment issues. Indeed, just three days after the Second Circuit issued its decision, another Federal court, the West Coast U.S. Court of Appeals for the Ninth Circuit, heard oral arguments in United States v. Harkonen, regarding InterMune Inc.’s public discussion of scientific trial experiments and their favorable results for the use of the company’s Actimmune drug for off-label purposes.4 On trial, the jury convicted InterMune’s CEO, W. Scott Harkonen, of devising a scheme to defraud and of making fraudulent statements regarding the drug’s efficacy. (The jury found this to be an unlawful wire fraud, but not a violation of the FDA’s “misbranding” policy.) Before the Ninth Circuit, Harkonen argued that the government’s prosecution of statements that were within the bounds of reasonable debate, and not unambiguously false or fraudulent, threatened to
stunt the evolution of scientific thought simply by suppressing expression of any view inconsistent with prevailing scientific orthodoxy. Such governmental power would conflict with all that the First Amendment has long been universally understood to represent.
But in March, the Ninth Circuit affirmed Harkonen’s conviction. The court sustained the trial jury’s findings that Harkonen’s statements were “fraudulent”, and that he had promoted the drug “with the specific intent to defraud[.]” This was sufficient to overcome Harkonen’s constitutional arguments, the court stressed, because “[t]he First Amendment does not protect fraudulent speech.”
s the Second Circuit recognized, Congress’s Food, Drug, and Cosmetic Act, as administered by the FDA, does not actually prohibit off-label uses of FDA-approved drugs; it merely attempts to stifle meaningful public discussion of those uses. To allow drugs to remain on the market for “off-label” uses yet stifle the “marketplace of ideas” regarding those drugs poses a direct threat to patients, as well as a challenge to the physicians’ Hippocratic Oath. As Dr. Scott Gottlieb has explained, the risk “is that the restraint on this communication is leaving doctors less rather than better informed, ultimately making patients more vulnerable.”5
Indeed, even the mere threat of prosecution has broad repercussions, chilling the very sort of speech that the government ought to be encouraging. As Gottlieb further explains, “today, some drug makers will not even distribute reprints of pivotal studies that appear in leading medical journals if those articles contain any mention of an off-label use of their products.”
Some of the FDA’s proponents argue that to oppose the FDA’s current efforts to criminalize discussion of off-label uses is to oppose FDA regulation altogether. The dissenting judge on the Second Circuit asserted that to follow the Second Circuit’s lead and defend free speech in this context “calls into question the very foundations of our century-old system of drug regulation.” But such dire warnings do not withstand scrutiny. First, the Second Circuit’s nuanced decision leaves the larger FDA regulatory framework entirely intact. The court did not rule that the FDA cannot prohibit the actual “misbranding” of drugs; the government can still prosecute companies that attempt to circumvent the FDA approval process. Just as important, it can still prosecute manifestly fraudulent speech, which has never enjoyed First Amendment protection, a fact that initial press reports did not make clear. And if there is gray area between what is manifestly fraudulent and what is not, the government can prosecute and a court can decide on a case-by-case basis. What the government can’t do, if the court’s decision stands, is criminalize truthful speech per se, as the government effectively attempted to do in the Caronia case when it focused almost exclusively on Caronia’s and Dr. Gleason’s remarks.
Indeed, the FDA itself has already acknowledged the narrow scope of the Second Circuit’s ruling. In the press release explaining its decision not to further litigate the Caronia case, the FDA agreed that
the decision does not strike down any provision of the FD&C Act or its implementing regulations, nor does it find a conflict between the Act’s misbranding provisions and the First Amendment or call into question the validity of the Act’s drug approval framework.
The regulatory structure remains sound; the FDA simply needs to stop implementing it in a way that criminalizes speech. Congress has not barred the off-label use of FDA-approved drugs—and for good reason—and so long as that is true the FDA must not criminalize discussion of such uses.
The Caronia case, and the broader public debate, highlights another important regulatory problem. If the FDA fears that “off-label” uses of FDA-approved drugs threaten to undermine the FDA approval process, it ought to improve the FDA approval process. If pharmaceutical companies could secure faster FDA review and approval of their drugs’ alternative uses, then doctors would not have to remain so reliant on off-label uses. The FDA was created to be a traffic cop, not a roadblock—to ensure that good drugs find their way to market, and bad drugs are kept out, but not to stifle the marketing of good drugs.
The difficulty that permeates this entire issue is the fact that corporate defendants face incredible pressure to immediately settle, and never fight back against, an FDA prosecution. The FDA can leverage the threat of “debarment”—a virtual corporate death penalty that prevents the company from doing business at all. Faced with that existential threat, corporate defendants readily submit to cutting a deal that will keep the company alive. (Recall that Orphan Medical promptly settled its Xyrem case with the FDA for $20 million.) What sets the Xyrem case apart is that the government made the mistake of targeting an employee who was ready to fight back and who was willing to risk personal liability for the sake of principle. Caronia’s refusal to acquiesce to the government’s charges is the only reason why the public can today benefit from the Second Circuit’s decision: Without Caronia, there would have been no United States v. Caronia, and none of the healthy debate that has ensued from that decision.
We need effective drugs on the market, and we need what is known in securities law as “truth on the market.” The Second Circuit demonstrated that both aims are achievable.
1Brandeis, Other People’s Money (1914), p. 92.
2David C. Riley, Stan N. Finkelstein and Randall Stafford, “Off-label Prescribing Among Office-Based Physicians”, Archives of Internal Medicine (May 2006).
3Government Opposition Brief at 9. The brief was formally directed at Dr. Gleason specifically, not Caronia, but because Caronia later joined Gleason’s motion the argument effectively applied to both of them.
4See Thomas Burton, “Courts to Weigh Free Speech Rights in Pharmaceutical Marketing Cases”, Wall Street Journal, December 6, 2012.
5Gottlieb, “The US Department of Justice’s Targeting of Medical Speech and its Public Health Impacts”, American Enterprise Institute (December 2012).