On September 6, 2012, the Ninth Circuit held that a plaintiff had failed to “adequately plead falsity” with respect to the reported results of a clinical drug trial in a Section 10(b) action where the plaintiff did “not allege that [the] [d]efendants misrepresented their own statistical methodology, analysis, and conclusions, but instead criticize[d] only the statistical methodology employed by [the] [d]efendants.” In re Rigel Pharmaceuticals, Inc. Sec. Litig., 2012 WL 3858112, at *9 (9th Cir. Sept. 6, 2012) (Hug, J.).
Rigel Pharmaceuticals is “a clinical-stage drug development company that discovers and develops novel, small-molecule drugs[.]” Id. at *1. “One of those drugs is R788, which Rigel is developing to treat and stop the progression of rheumatoid arthritis.” Id.
On December 13, 2007, Rigel issued a press release discussing the results of its Phase 2 clinical study of R788. The press release reported that R788 had “‘demonstrated statistically significant results’” in the treatment of rheumatoid arthritis. Id. at *2. With respect to side effects, the press release stated that “‘[t]he most common clinically meaningful adverse events noted in the clinical trial were dose-related neutropenia, mild elevations of liver function tests, and gastrointestinal (GI) side effects.’” Id.
Rigel presented “much more extensive, detailed, and scientific information” about the R788 Phase 2 clinical trial in an article published in the November 2008 edition of Arthritis and Rheumatism, a medical journal. Id. at *4. “[I]n addition to the more severe adverse events disclosed in the original reports,” the journal article noted that several patients had experienced other side effects such as “smaller elevations of liver enzymes” and “mild hypertension[.]” Id.
A plaintiff subsequently brought suit alleging that the results of Rigel’s Phase 2 clinical trial for R788 were “false” because Rigel had used “statistically false p-values” and had relied on “inaccurate and improper statistical analysis.” Id. at *7 (internal quotations omitted). The plaintiff also contended that the December 2007 press release was misleading insofar as it failed to provide a complete list of R788 side effects. The plaintiff asserted claims under Sections 10(b) and Rule 10b-5, among others. In August 2010, the district court dismissed the complaint, holding, inter alia, that “disagreements over statistical methodology and study design are insufficient to allege a materially false statement.” Id.
The Ninth Circuit Holds That Using the “Wrong” Statistical Methodology in Compiling and Reporting Drug Trial Data Does Not Render That Data False or Misleading
On appeal, the Ninth Circuit held that the plaintiff’s allegations were “not about false statements” but instead “concern[ed] two different judgments about the appropriate statistical methodology to be used by [the] [d]efendants.” Id. The plaintiff “did not allege that [the] [d]efendants [had] inaccurately reported the results of their own statistical analysis[,]” or that they “had chosen or changed their statistical methodology after seeing the unblinded raw data from the clinical trial.” Id. Rather, the plaintiff’s “allegations of ‘falsity’ essentially [were] disagreements with the statistical methodology adopted by the doctors and scientists who designed and conducted the study, wrote the journal article, and selected the article for publication.” Id.
The plaintiff contended that it was “simply challenging the truth of the reported results, not the study design[.]” Id. The Ninth Circuit found “multiple problems with this argument.” Id. First, the court emphasized that the plaintiff was “alleging that [the] [d]efendants should have used different statistical methodologies, not that [the] [d]efendants [had] misrepresented the results they obtained from the methodologies they employed.” Id. Second, the Ninth Circuit explained that accepting the plaintiff’s argument would require it “to draw a line between using a particular method of statistical analysis that was part of a study’s protocol … and disclosing results that were calculated using that statistical analysis.” Id. “Drawing such a distinction would suggest that a company should announce statistical results that are obtained using a statistical methodology that is adopted after the study data is made available to the researchers and that is different from the methodology used as part of the clinical trial.” Id. “Such a post-hoc adoption of a statistical method could raise concerns regarding reliability, biased scientific methods, or even fraud.” Id.4
The Ninth Circuit noted that “[n]either the Supreme Court nor [the Ninth Circuit] has addressed the question of whether statements concerning statistical results of a clinical trial may be considered false or misleading under Rule 10b-5 because the statistical methodology that produced those results was not the best or most acceptable methodology.” Id. at *8. However, other district courts to consider the issue have held that “merely alleging that defendants should have used different statistical methodology in their drug trials is not sufficient to allege falsity.” Id. For example, in Padnes v. Scios Nova Inc., 1996 WL 539711 (N.D. Cal. Sept. 18, 1996) (Patel, J.), the court determined that “the fact that the plaintiffs disagreed with the researchers about the import of the data did not make the defendants’ summaries of the study false or misleading.” In re Rigel, 2012 WL 3858112, at *8. The Padnes court “concluded that the securities laws do not require that companies report information only from optimal studies” and held that “companies reporting information from imperfect studies are not required to disclose alternative methods for interpreting the data.” Id.5
Finding this reasoning “persuasive,” the Ninth Circuit held that the plaintiff “did not adequately plead falsity with respect to [the] statistic results” of the R788 clinical trial. Id. at *9.
The Ninth Circuit Determines Rigel Had No Duty to Disclose More Information Concerning R788’s Side Effects in the December 2007 Press Release
The Ninth Circuit found that the “subsequent release of more extensive information” concerning the side effects of R788 in a medical journal article did not render the statements in the December 2007 press release false or misleading because the later-disclosed information “was not inconsistent with the results that originally were reported.” Id. at *10.
The plaintiff relied on the Supreme Court’s decision in Matrixx Initiatives, Inc. v. Siracusano, 131 S. Ct. 1309 (2011) (Sotomayor, J.) to argue that “once a company chooses to disclose any safety information, it must disclose all material information regarding safety.” In re Rigel, 2012 WL 3858112, at *14 n.8. The Ninth Circuit held that “[t]his contention misconstrues” Matrixx. Id. “The Matrixx Court made it clear that not all adverse events would be material and, more importantly, that not all material adverse events would have to be disclosed.” Id. “Thus, as long as the omissions do not make the actual statements misleading, a company is not required to disclose every safety-related result from a clinical trial, even if the company discloses some safety-related results and even if investors would consider the omitted information significant.” Id.