Welcome
Pharmaceutical & Life Sciences News

Room for Debate:
First Amendment Protection for Public Disclosures of Clinical Trials

Dec. 16, 2011, 5:00 AM

Navigating the challenges faced by public companies making public disclosures in the context of a clinical trial can be difficult. It is not always clear what and how information should be disclosed to adhere to boundaries created by securities laws, as well as valid business and investor objectives. Questions concerning the materiality of clinical data and results, and the conclusions that might be drawn therefrom, are often debatable. Potential pitfalls of both overdisclosure and underdisclosure must be avoided.1See generally, Geoffrey M. Raux and J. Mark Waxman, “Navigating Public Disclosures During Drug Development,” 4 BNA Life Sciences Law & Industry Report 1101, 11/19/10. The prospect of criminal prosecution, however, catapults the risks to another level. It also raises questions as to what degree the government should be permitted to regulate and prosecute individuals or companies on the basis of disclosures (i.e., a form of speech) in the clinical trial context where objective truth is hard to define.

The potential for First Amendment concerns in this context is exhibited in the prosecution, conviction, and recent appeal of InterMune Inc.’s former chief executive officer, Dr. W. Scott Harkonen. Dr. Harkonen’s prosecution has centered on statements he made in a press release concerning one of the company’s phase III clinical trials. The case is now before the Ninth Circuit Court of Appeals, and has caught the attention of many in the pharmaceutical community, including Pharmaceutical Research and Manufacturers of America, the trade association representing the nation’s leading research-based pharmaceutical and biotechnology companies. PhRMA has joined the case as amicus curiae seeking to overturn Dr. Harkonen’s conviction on First Amendment grounds.

The appeal largely turns on whether the statements in the press release were valid expressions of a scientific opinion that should have been protected from criminal prosecution by the First Amendment, or whether they were simply statements of a commercial nature that, as the jury concluded, amounted to intentional fraud. In deciding the appeal, the Ninth Circuit faces difficult and potentially far-reaching questions, including the precise nature of speech in the clinical trial context and what First Amendment protections are necessary to ensure that improper convictions on the basis of such speech are avoided. If the court seeks to resolve these questions, its decision may have widespread implications for companies and individuals involved with clinical trials and how they interact with the public in making their disclosures.

Dr. Harkonen’s Criminal Prosecution

In October 2006, InterMune entered a deferred prosecution agreement with the government and agreed to pay a $37 million penalty to resolve charges with respect to illegal off-label marketing of its drug, Actimmune, and for false claims for reimbursement under the Medicare program.2See DOJ Press Release #06-728, Biopharmaceutical Firm Intermune to Pay U.S. Over $36 Million for Illegal Promotion and Marketing of Drug Actimmune (Oct. 26, 2006). Subsequently, in March 2008, the grand jury indicted Dr. Harkonen, InterMune’s former CEO, on charges of wire fraud and misbranding in violation of 18 U.S.C. §1343 and 21 U.S.C. §§331(k), 333(a)(2) and 352(a) respectively.3See DOJ Press Release #08-217, Former Intermune CEO W. Scott Harkonen Indicted for Wire Fraud and FDA Violations (March 18, 2008). The prosecution of Dr. Harkonen centered on the disclosure of phase III clinical trial results made by Dr. Harkonen with respect to Actimmune.4See United States v. Harkonen, No. 08-00164, 2010 WL 2985257, at *1 (N.D. Cal. July 27, 2010). The phase III trial had failed to meet its primary end point, but unblinded subsequent analyses indicated a potential benefit to patients suffering from mild to moderate cases of idiopathic pulmonary fibrosis (IPF).5United States v. Harkonen, No. 08-00164, 2009 WL 158712, at *2 (N.D. Cal. June 4, 2009). The FDA, however, informed InterMune that the trial data were not sufficient to gain approval for treatment of IPF, and that—notwithstanding the subsequent analyses—further clinical testing would be required.6Id.

Despite the FDA’s position, on Aug. 28, 2002, InterMune issued a positive press release of the trial results.7Id. The release announced that trial data demonstrated a survival benefit for patients suffering from mild to moderate IPF, reducing mortality by 70 percent.8Id. The release was distributed by e-mail to InterMune sales representatives and a fax of the release went to more than 2,000 pulmonologists (physicians who treat IPF).9Id. Dr. Harkonen also directed that T-shirts be distributed to sales staff and other employees celebrating the announcement of the trial results.10Id. The T-shirts stated the name of the Actimmune trial on the front, and read “Feel Better Live Longer” on the back.11Id. Although the press release indicated that the trial’s primary end point did not reach statistical significance, it did not reveal that the conclusions for the subgroup of mild to moderate IPF were taken from retrospective analyses performed after data were unblinded.12 See Harkonen, 2010 WL at *9-10. During 2002 and 2003, sales for Actimmune dramatically increased, primarily as a result of off-label prescriptions for treating IPF.13See Harkonen, 2009 WL at *3.

Subsequent to the release, InterMune developed a follow-on trial that commenced in December 2003. By that time, Dr. Harkonen had resigned his position with the company. The follow-on trial continued through 2007, when it ultimately was abandoned after results demonstrated that Actimmune provided no actual benefits to patients with IPF.14Id. at *2.

On March 23, 2009, in response to his indictment, Dr. Harkonen filed a motion to dismiss arguing that the statements included in the press release should be excluded as evidence of his culpability at trial, as these statements represented “scientific speech about ‘medical practices in fields where knowledge ha[d] not yet been crystallized in the crucible of experience’ and where there exist[ed] ‘no exact standard of absolute truth by which to prove the assertions false and a fraud.’”15Id. at *4. The court rejected these arguments, holding that Dr. Harkonen could not “successfully argue that ‘imperfect knowledge’ in the field somehow sanitized the press release’s communication that the clinical trial data, albeit missing its primary endpoint [and being rejected by the FDA], suggested a mortality benefit in a subgroup of IPF patients.”16Id. at *7. The case proceeded to trial, and, on Sept. 29, 2009, the jury handed down a guilty verdict on the wire fraud count.17See Harkonen, 2010 WL at *1.

Dr. Harkonen was sentenced in April 2011 to six months of home confinement and fined $20,000 for his role in disseminating misleading information about Actimmune.18See Karen Gullo, Intermune Ex-CEO Harkonen Sentenced to Home Confinement in Fraud Case, Bloomberg (April 14, 2011). Notably, the government had sought a sentence of 10 years and a fine of $1 million, but the district court refused to impose such a harsh sentence, noting that there was little evidence of loss and that the court was unable to “figure out who a victim is in this case, and whether the victims were benefited in some way at all or not.”19See App. Br. of Dr. W. Scott Harkonen (“Harkonen App.”) at 30-32 (Oct. 28, 2011).

The First Amendment Arguments on Appeal

On appeal, Dr. Harkonen has asserted a number of arguments, the most interesting of which concern the relationship between the First Amendment and the government’s ability to prosecute individuals for statements made in public disclosures in the clinical trial context. These First Amendment arguments have been picked up and augmented by PhRMA. Together, Dr. Harkonen and PhRMA characterize the statements in the press release as expressions of scientific opinion on a subject matter over which reasonable scientists could disagree.20See App. Br. of PhRMA (“PhRMA Br.”) at 3 (Nov. 4, 2011). They argue that affirming his conviction “would threaten core First Amendment principles by establishing that scientific debate over how to interpret data can constitute a crime.”21Id. at 2. The prosecution and conviction are of “great concern,” moreover, due to the chilling effect they might have on others. PhRMA points out that “pharmaceutical companies constantly communicate with the public about complex scientific and medical issues that are the subject of ongoing and good-faith debate [and] [a] robust dialogue on these matters helps to foster the innovation that leads to advancements in medicine and better outcomes for patients.”22Id. PhRMA contends that this “discourse would be seriously chilled if [Dr. Harkonen’s conviction] is affirmed.”23Id.

According to Dr. Harkonen and PhRMA, even if Dr. Harkonen’s statements in the press release represented a minority position, it should not be considered a fraudulent one subject to criminal prosecution. PhRMA attempts to draw a distinction between “fabrication of data or other outright misstatements of objectively verifiable and material facts,” from “[m]atters of interpretation that can be fairly debated by reasonable scientists and are not empirically verifiable.”24Id. at 11. PhRMA contends that statements in the latter category should not be sent “to the jury unless the level of scientific consensus is such that no reasonable expert could find the defendant’s statement to be true.”25Id. In the case of the press release, Dr. Harkonen and PhRMA argue that the statements—which they characterize as debatable interpretations of data about the potential efficacy of Actimmune for treating IPF—should be protected by the First Amendment as scientific expression and debate critical to spurring medical progress, and ultimately improving patient care and saving lives.26See id. at 2-19; Harkonen Br. at 43-53.

The district court rejected these arguments when advanced in the initial motion to dismiss, and in the post-conviction motions. The court found that “[t]he mere fact that Harkonen is an M.D., that the press release he prepared presented actual data and statistical analyses, and that the dissemination of the press release may have generated vigorous debate in the pulmonological and pharmaceutical analyst community, do not disturb th[e] conclusion that his statements were not protected by the First Amendment.”27See Harkonen, 2009 WL at *6. The court noted that the government was not simply targeting Dr. Harkonen’s speech because of its subject matter or scientific viewpoint, but that the government had alleged that his statements were fraudulent and made with a specific intent to defraud others.28Id. at *4-6. As the court stressed, the First Amendment does not protect fraud.29See Harkonen, 2010 WL at *15. Because Dr. Harkonen’s statements were not per se protected by the First Amendment, it was permissible to send the case to the jury for a determination of whether the government had demonstrated fraud beyond a reasonable doubt.30See Harkonen, 2009 WL at *4-6. In considering the guilty verdict, the court found that the jury had been presented with sufficient evidence and had concluded that the statements made by Dr. Harkonen were indeed fraudulent, that he knew they were fraudulent, and that he made them with the specific intent to defraud others.31See Harkonen, 2010 WL at *3-14.

Scientific Speech or Commercial Promotion?

Beneath the competing positions of Dr. Harkonen/PhRMA and the district court are different conceptions of the speech at issue in the press release. The conception of the speech is critical, because the Supreme Court has drawn a distinction between “speech proposing a commercial transaction, which occurs in an area traditionally subject to government regulation, and other varieties of speech.”32Bolger v. Youngs Drug Prods. Corp., 463 U.S. 60, 64 (1983). So-called “commercial speech,” which the court has characterized as “speech which does no more than propose a commercial transaction,” is afforded only limited protection under the First Amendment.33Id. at 66. On the other hand, rigorous First Amendment protection is afforded to noncommercial speech, including speech involving “scientific expression and debate.”34Bd. of Trustees of Leland Stanford Junior Univ. v. Sullivan, 773 F. Supp. 472, 474 (D.D.C. 1991). The Supreme Court has held that “[t]he First amendment protects [speech] which, taken as a whole, ha[s] serious literary, artistic, political, or scientific value, regardless of whether the government or a majority of the people approve of the ideas [the speech] represent[s].”35Miller v. California, 413 U.S. 15, 34 (1973).

For Dr. Harkonen and PhRMA, the press release involved noncommercial speech of scientific value, and, therefore, should have been entitled to per se protection under the First Amendment. Dr. Harkonen and PhRMA argue that the speech “could easily be found in a medical or scientific journal, or in a debate among colleagues in an academic setting,” and was, therefore, “unlike speech … that commonly appears in more traditional advertising or labeling contexts.”36PhRMA Br. at 21-22. Analyzed in this framework, the government should have been completely barred from prosecuting Dr. Harkonen on the basis of the press release.

The district court viewed the speech quite differently, finding it to be purely commercial. The district court found “that [because] the speech [was] a press release and not a peer-reviewed publication, that it refer[red] to a specific commercial product on the market … , and that it was unquestionably disseminated for commercial benefit … [took it] outside the realm of pure science speech and move[d] it towards the realm of commercial speech,” entitling it to only “qualified” protection.37Harkonen, 2009 WL at *6. Because the court saw the press release as commercial speech, the First Amendment scrutiny it applied was less rigorous than it otherwise might have been.

The precise nature of Dr. Harkonen’s speech—as well as that of other clinical trial disclosures—is probably somewhere in the middle of these respective conceptions. Indeed, clinical trial disclosures serve scientific, commercial, and regulatory aims. They describe analyses from scientific data and advance scientific conclusions—a scientific aim. They promote key products in development for drug companies, often overemphasizing the positive results while playing down the negatives—a commercial aim. And they also serve to inform investors and the public as to the company’s development progress in compliance with applicable laws—a regulatory aim. Such multifaceted speech cannot be pigeon-holed as purely scientific, or purely commercial. Speech of such a mixed nature presents “thorny” issues in terms of the First Amendment concerns involved.38Id. (noting that “the case law [is] still in an unsettled state … [and] questions remain about when such ‘pure’ [scientific] speech gets converted to a ‘less pure’ form of commercial speech when a drug company is involved”). While the concerns are probably not as pronounced as Dr. Harkonen and PhRMA contend, they do appear to be graver than the district court considered them, demanding sufficient protections.

What Protections Are Required?

Dr. Harkonen and PhRMA raise legitimate concerns over the chilling effect criminal prosecution might have on genuine scientific debate. Clinical trials inevitably involve interpretation and extrapolation of clinical data and results, in an effort to determine potential drug efficacy. The conclusions that can be derived from these types of data (often preliminary in nature) are not always clear-cut, and it is through rigorous debate and competing analyses that progress is spurred. Increased risk of criminal prosecution certainly could quell this progress, and the government and courts should be wary before taking overly aggressive action.

To protect against this chilling effect, Dr. Harkonen and PhRMA argue that criminal prosecution of speech in the clinical trial context should only be permitted if “no reasonable expert could find the … statement to be true.” If a reasonable expert can be found to affirm the statement, Dr. Harkonen and PhRMA contend that criminal prosecution should be barred. This would almost certainly eliminate any chilling effect on genuine scientific debate, but it seems overly restrictive and ultimately unjustified. The mere fact that an expert could be proffered in support of a statement made in a disclosure should not prohibit the government from presenting contrary evidence that the speech was false, that the speaker actually knew it was false, and that he had the specific intent to defraud others when making it. There still should be room for the prosecution and conviction of individuals or companies engaged in knowing and intentional fraud, which, as the district court stressed, is not protected by the First Amendment. Simply because an act of fraud occurs in a field that is important to medical and scientific development, or concerns matters of scientific debate, does not mean that it should be immune from criminal prosecution.

Maintaining the ability to prosecute, therefore, while avoiding the negative impact on scientific speech and progress that might result from overzealous prosecution is a delicate, but necessary balance. The district court saw this balance satisfied in the government’s burden of proof, which required it to demonstrate, beyond a reasonable doubt, that Dr. Harkonen knowingly engaged in fraud and had the specific intent to deceive. In the court’s opinion, this burden provided sufficient protection against any First Amendment concerns.

The difficulty with the court’s position is that the government’s ultimate burden of proof does not really come into play until the door to prosecution has been opened. The burden is primarily implicated when the case is at its conclusion—after years of litigation and the specter of a jail sentence pending—at the time of the jury’s verdict and in consideration of post-trial motions. At the front end, when deciding whether prosecution should be permitted in the first place, these standards do not have the same force. In fact, Dr. Harkonen’s pre-trial motion to dismiss was rejected, in large part, because he was “hamstrung” by the requirement that the court must accept all allegations in the indictment as true.39See id. at *7. In other words, because the indictment characterized the statements at issue as knowing and intentional fraud, the court had to analyze the First Amendment concerns with this characterization as a working assumption.

Although the government’s burden of proof becomes more applicable at the back end—with motions for a directed verdict, with respect to the jury’s decision, and with post-trial motions challenging the jury’s decision—it still appears insufficient to ensure that only truly fraudulent conduct results in convictions, and good faith expressions of minority or debatable scientific opinions do not. The insufficiency of the government’s burden as the sole protection against improper convictions in this context stems from the type of evidence that the government might use to meet its burden. To prove the falsity of a statement, for example, the government will likely introduce testimony from witnesses holding a contrary viewpoint—just as it did in the Harkonen case. Even if the defense presents competing experts, a jury without scientific training will effectively be forced to decide whether a challenged viewpoint is a minority position or an outright falsity. This decision, in turn, will almost certainly be impacted by the government’s evidence of fraudulent intent, which will likely center on the clear profit motive common to all commercial drug development. Recognizing this profit motive, it is not hard to imagine a jury of laypersons finding that a minority scientific position is knowing and intentional fraud.

Additional protections—other than the government’s burden of proof—do, therefore, appear to be warranted. These additional protections could be provided in the form of mandatory jury instructions that would reduce the likelihood of a jury convicting minority scientific opinions. Jury instructions on good faith, puffery, and the difference between minority opinions and scientific falsehood would be appropriate to reach an adequate balance.40Dr. Harkonen expressly requested instructions on good faith and puffery at trial, but these requests were denied by the district court. The court’s refusal to give these instructions is one of the bases on which Dr. Harkonen now seeks reversal. See Harkonen Br. at 68-71. A mandatory good faith instruction would make clear that a scientific position honestly held, even if it turned out to be inaccurate, incorrect, or wrong, cannot be punishable as fraud. A puffery instruction could be used to inform the jury that statements that simply magnify the advantages of a product, without falsely asserting the existence of qualities the product does not possess, are not to be considered fraudulent. Finally, an instruction distinguishing between minority or debatable scientific opinion and outright falsehood could serve to remind the jury that convictions should not be handed down for reasonable views held in good faith, even if those views conflict with the general or majority consensus. These instructions would not only make a conviction of a minority or debatable scientific position less likely, they could dissuade the government from prosecuting cases in the gray area, where this enhanced burden would be too difficult to overcome.41Whether the courts ultimately adopt these or other safeguards for criminal prosecution involving clinical trial disclosures, it is important to remember that such safeguards would not restrict the ability of the government to initiate criminal prosecution in the first instance, and would have no bearing on the civil liability context. The United States Supreme Court’s recent decision in Matrixx Initiatives Inc. v. Siracusano, __ U.S. __, 131 S.Ct. 1309 (2011) reaffirms the relatively low bar for initiating fraud actions involving public disclosures. In Matrixx, the Supreme Court addressed the appropriate standard for disclosure in the context of a securities fraud class action involving the alleged failure to disclose a possible link between the drug in question (Zicam Cold Remedy) and loss of smell, thereby rendering statements made by Matrixx misleading. The principal issue was whether an allegation (and ultimately evidence) of statistical significance was necessary to assert a claim. Citing its recent decision in Basic Inc. v. Levinson, 485 U. S. 224, 231-232 (2011), the court rejected any “bright-line rule,” characterizing the relevant standard as whether there was “a substantial likelihood that the disclosure of the omitted fact would have been viewed by the reasonable investor as having significantly altered the total mix of information made available” (Matrixx, 131 S.Ct. at 1318-19). The court found that “assessing the materiality of [information] is a ‘fact-specific’ inquiry that requires consideration of the source, content, and context of the [information] … [t]hat is not to say that statistical significance (or lack thereof) is irrelevant—only that it is not dispositive of every case” (Id. at 1321).

Conclusion

Whether the Ninth Circuit reverses Dr. Harkonen’s conviction on improper jury instructions or on some other ground, or if it simply affirms his conviction, its decision will be closely watched by the clinical trial community and may have a significant impact on the degree of First Amendment protection that is afforded to clinical trial disclosures going forward. In the meantime, the Harkonen appeal is another reminder of the importance of making accurate and thorough disclosures during the course of clinical trials. Whatever the Ninth Circuit decides, the possibility of criminal prosecution should cause all pharmaceutical companies to re-evaluate their policies and procedures for issuing public disclosures and ensure that these disclosures are given the attention they require.