In Re: Merck & Co ( 2008 )

Opinions of the United
2008 Decisions States Court of Appeals
for the Third Circuit
9-9-2008
In Re: Merck & Co
Precedential or Non-Precedential: Precedential
Docket No. 07-2431
Follow this and additional works at: http://digitalcommons.law.villanova.edu/thirdcircuit_2008
Recommended Citation
"In Re: Merck & Co " (2008). 2008 Decisions. Paper 451.
http://digitalcommons.law.villanova.edu/thirdcircuit_2008/451
This decision is brought to you for free and open access by the Opinions of the United States Court of Appeals for the Third Circuit at Villanova
University School of Law Digital Repository. It has been accepted for inclusion in 2008 Decisions by an authorized administrator of Villanova
University School of Law Digital Repository. For more information, please contact Benjamin.Carlson@law.villanova.edu.
PRECEDENTIAL
UNITED STATES COURT OF APPEALS
FOR THE THIRD CIRCUIT
Nos. 07-2431, 07-2432
IN RE: MERCK & CO., INC. SECURITIES,
DERIVATIVE & “ERISA” LITIGATION
(MDL No. 1658)
CONSOLIDATED SECURITIES LITIGATION
Richard Reynolds, Steven LeVan, Jerome Haber and
The Public Employees’ Retirement System of Mississippi,
the Court-Appointed Lead Plaintiffs and Plaintiffs Union
Asset Management Holding AG, Loren Arnoff,
Robert Edwin Burns, Jan Charles Finance S.A.,
Martin Mason, Frank H. Saccone, Charlotte Savarese,
Joe Savarese, Joseph Goldman, Sherrie B. Knuth,
Joseph S. Fisher, M.D., Naomi Raphael,
Rhoda Kanter, Park East, Inc. and Marc Nathanson, on behalf
of themselves and the proposed class of purchasers of Merck
securities during the period between
May 21, 1999 and October 29, 2004,
Appellants
On Appeal from the United States District Court
for the District of New Jersey
(D.C. Nos. 05-cv-01151, 05-cv-02367)
District Judge: Honorable Stanley R. Chesler
Argued June 24, 2008
Before: SLOVITER, BARRY, and, ROTH Circuit Judges
(Filed September 9, 2008)
______
John P. Coffey (Argued)
Bernstein, Litowitz, Berger & Grossman
New York, NY 10019-0000
Bruce D. Bernstein
New York, NY 10022-0000
Paul B. Brickfield
Brickfield & Donahue
River Edge, NJ 07661-0000
David A.P. Brower
Brower Piven
New York, NY 10022-0000
James E. Cecchi
Lindsey H. Taylor
Carella, Byrne, Bain, Gilfillan, Cecchi,
Stewart & Olstein
Roseland, NJ 07068-0000
Mark Levine
Stull, Stull & Brody
New York, NY 10017-0000
Michael Miarmi
New York, NY 10119-0165
Lee A. Weiss
New York, NY 10022-0000
Richard H. Weiss
New York, NY 10119-0165
Robert T. Haefele
Motley Rice
Mount Pleasant, SC 29465-0000
2
Attorneys for Appellants
Robert H. Baron
Evan R. Chesler (Argued)
Cravath, Swaine & Moore
New York, NY 10019-0000
Roberta Koss
William R. Stein
Hughes, Hubbard & Reed
Washington, DC 20006-0000
William H. Gussman
Martin L. Perschetz
Sung-Hee Suh
Schulte, Roth & Zabel
New York, NY 10022-0000
Lawrence M. Rolnick
Sheila A. Sadighi
Lowenstein Sandler
Roseland, NJ 07068-0000
Attorneys for Appellees
_____
OPINION OF THE COURT
SLOVITER, Circuit Judge.
Appellants, purchasers of Merck & Co., Inc. stock, filed the
first of several class action securities fraud complaints on
November 6, 2003, alleging that the company and certain of its
officers and directors (collectively, “Merck”) misrepresented the
safety profile and commercial viability of Vioxx, a pain reliever
that was withdrawn from the market in September 2004 due to
safety concerns. The District Court granted Merck’s motion to
dismiss the complaint under Rule 12(b)(6) of the Federal Rules of
Civil Procedure, holding that Appellants were put on inquiry notice
3
of the alleged fraud more than two years before they filed suit, and
thus their claims were barred by the statute of limitations.
Appellants argue that the District Court erred in finding as a matter
of law that there was sufficient public information prior to
November 6, 2001 to trigger Appellants’ duty to investigate the
alleged fraud. Because the District Court dismissed on the basis of
the complaint, we must accept its allegations as true.1
I.
Factual Background
In May 1999, the Food and Drug Administration (“FDA”)
approved Vioxx, a new drug introduced by the pharmaceutical
company Merck. Vioxx is the brand name of rofecoxib, a
nonsteroidal anti-inflammatory drug (“NSAID”) used in the
treatment of arthritis and other acute pain. Most NSAIDs, such as
aspirin, ibuprofen, and naproxen, function by inhibiting two
enzymes: cyclooxygenase-1 (“COX-1”), which is associated with
the maintenance of gastrointestinal (“GI”) mucus and platelet
aggregation, and cyclooxygenase-2 (“COX-2”), which is associated
with the response to pain and inflammation. The inhibition of
COX-1 leads to harmful GI side effects. Because Vioxx was
designed to suppress COX-2 without affecting COX-1, Merck
marketed Vioxx as possessing the beneficial effects of traditional
NSAIDs but without the harmful GI side effects associated with
those drugs. The market viewed Vioxx as a potential “blockbuster”
drug for the company, App. at 469, and as its “savior,” App. at 494.
1
The District Court took judicial notice of the various
public documents submitted to it in connection with the motion to
dismiss. Appellants do not challenge this decision on appeal and
we see no reason to disturb it. “The inquiry notice analysis is an
objective one. Whether appellants read the [documents] or were
aware of them is immaterial. They serve only to indicate what was
in the public realm at the time, not whether the contents of those
[documents] were in fact true.” Benak ex rel. Alliance Premier
Growth Fund v. Alliance Capital Mgmt., L.P., 435 F.3d 396, 401
n.15 (3d Cir. 2006).
4
Merck repeatedly touted the safety profile, sales, and commercial
prospects of the drug in press releases, public statements, and
Securities and Exchange Commission (“SEC”) filings throughout
the class period.
A. Pre-FDA Approval and the VIGOR Study (1996 -
March 2000)
Prior to the FDA’s approval of Vioxx, officials at Merck
were concerned that Vioxx could cause harmful cardiovascular
(“CV”) events, such as heart attacks. Internal emails from 1996
and 1997 demonstrate that Merck employees were aware that there
was “a substantial chance” and a “possibility” of CV events that
could “kill [the] drug.” App. at 496. In 1998, an unpublished
internal Merck clinical trial entitled Study 090 revealed that Vioxx
caused a greater incidence of CV events than a placebo or a
different arthritis drug.2
In January 1999, Merck commenced the VIOXX
Gastrointestinal Outcomes Research (“VIGOR”) study, which
compared Vioxx to naproxen, the active ingredient in brand-name
pain relievers such as Aleve and Naprosyn.3 Although the study
showed that Vioxx had a GI safety profile superior to that of
naproxen, it also showed that Vioxx users had a higher incidence
of CV events than naproxen users. In a March 9, 2000 email,
defendant Edward Scolnick, the President of Merck Research
Laboratories, acknowledged the existence of CV events,
2
The sources upon which Appellants rely in making these
allegations were first made public in November 2004,
approximately a year after Appellants filed their initial complaint.
See Anna Wilde Mathews & Barbara Martinez, Warning Signs:
E-Mails Suggest Merck Knew Vioxx’s Dangers at Early Stage,
Wall St. J., Nov. 1, 2004, at A1; 60 Minutes (CBS television
broadcast Nov. 14, 2004) (transcript available on LexisNexis).
3
See ALEVE FAQs, http://www.aleve.com/faqs.html#g21
(last visited July 25, 2008); Roche Pharmaceuticals in the U.S., Our
Products, Naprosyn, http://www.rocheusa.com/products/naprosyn/
(last visited July 25, 2008).
5
commenting, “it is a shame but it is a low incidence and it is
mechanism based as we worried it was.” App. at 512.
Merck did not attempt to conceal the results of the VIGOR
study. It made them public in a press release on March 27, 2000,
that emphasized Vioxx’s superior GI safety profile but also noted
the incidence of CV events. Merck stated:
[S]ignificantly fewer thromboembolic events were observed
in patients taking naproxen in this GI outcomes study,
which is consistent with naproxen’s ability to block platelet
aggregation. This effect on these events had not been
observed previously in any clinical studies for naproxen.
Vioxx, like all COX-2 selective medicines, does not block
platelet aggregation and therefore would not be expected to
have similar effects.
App. at 765. The press release also stated that “[a]n extensive
review of safety data from all other completed and ongoing clinical
trials, as well as the post-marketing experience with Vioxx, showed
no indication of a difference in the incidence of thromboembolic
events between Vioxx, placebo and comparator NSAIDs.” App. at
766.
The VIGOR study results were widely reported in the press,
medical journals, and securities analyst reports. Market analysts
and members of the press immediately understood that CV events
could be a side effect of Vioxx. Nonetheless, many observers also
took notice of Merck’s hypothesis that naproxen lowered CV
events (the “naproxen hypothesis”). The naproxen hypothesis
attributed the results of the VIGOR study to the beneficial effects
of naproxen’s blocking of platelet aggregation rather than to the
harmful effects of Vioxx in causing thromboembolic events. The
issue whether naproxen lowered the heart attack risk or Vioxx
caused it was thus presented. While many analysts noted that the
naproxen hypothesis was unproven, some also concluded that it
was the most likely explanation for the increased CV events
observed in the VIGOR study.
One representative article distributed by Reuters on April
6
27, 2000, quoted a Merck spokesman who acknowledged the
“statistically significant” finding that patients of Vioxx had a
higher rate of CV events, but suggested that this might be
explained by a beneficial effect of naproxen. App. at 2287. In that
same article, however, a spokesperson for the manufacturer of
Naprosyn explained that the company had no knowledge that
naproxen prevented heart attacks or strokes; similarly, an analyst
for ABN Amro suggested that he was skeptical of Merck’s
explanation.
B. FDA AAC Hearing (February 8, 2001)
On February 8, 2001, the FDA’s Arthritis Advisory
Committee (“AAC”) held a public hearing to consider Merck’s
request to include the positive GI results from the VIGOR study in
its Vioxx labeling. Six days before that hearing, J.P. Morgan
issued a research report summing up the state of knowledge about
Vioxx after the VIGOR study. The report stated that the basic idea
behind the naproxen hypothesis was “poorly proven,” and that
there was “no way to retrospectively slice the data to prove the
NSAID benefit vs. Vioxx risk argument,” although one existing
theory “might support a ‘Vioxx risk’ hypothesis.” App. at 2547.
J.P. Morgan warned, “[t]his is the type of clinical ‘signal’ that was
ignored, and later haunted the FDA in recent drug recalls like
Warner Lambert’s Rezulin and Glaxo’s Lotronex.” App. at 2547.
During the AAC hearing, defendant Alise Reicin, Executive
Director of Clinical Research at Merck Research Laboratories,
explained to the panel, “when you review the results of VIGOR in
isolation you don’t know whether the imbalance of cardiovascular
events was caused by a decrease in events on a platelet-inhibiting
NSAID, naproxen, or an increase in events on a COX-2 selective
inhibitor,” i.e., Vioxx. App. at 995. She then suggested that
naproxen was likely responsible for the difference in CV events
observed in users of the two drugs. At the public portion of the
hearing, the panel subsequently discussed whether to call for the
inclusion of a warning in the Vioxx labeling stating that it was
“uncertain” whether the CV events noticed in VIGOR were “due
to beneficial cardioprotective effects of naproxen or prothrombotic
effects of [Vioxx], and leave it at that, that basically we don’t know
7
the reason.” App. at 1143.
Nonetheless, some press accounts reported that certain AAC
panel members asserted that “[d]ifferences in cardiac risk between
Vioxx and naproxen appeared to result from a beneficial effect of
naproxen, not a danger from Vioxx,” App. at 2311, and that there
was “some reassurance that what we see, in effect, is a protective
effect of naproxen,” App. at 2306. In subsequent coverage, many
securities analysts reported that the hearing had benefited Merck
and they continued to project substantial future revenues for Vioxx.
However, at least one investment firm issued a report stating, “our
skepticism relating to naproxen having a cardioprotective effect is
reinforced” by the AAC hearing. App. at 2703.
C. First Vioxx Product Liability Lawsuit (May 2001)
In May 2001, a product liability lawsuit was filed jointly
against Merck and the makers of Celebrex, a rival COX-2 selective
inhibitor. The complaint alleged that the pharmaceutical
companies “have consistently marketed Vioxx and Celebrex as
highly effective pain relief drugs for patients suffering from
osteoarthritis,” despite the fact that “Merck’s own research”
demonstrated that “users of Vioxx were four times as likely to
suffer heart attacks as compared to other less expensive
medications, or combinations thereof.” App. at 1748. The
plaintiffs sought “emergency notice to class members and revised
patient warnings, in the form of additional medical labeling which
is presently being considered by the FDA . . . .” App. at 1748.
D. JAMA Article (August 22, 2001)
On August 22, 2001, the Journal of the American Medical
Association (“JAMA”) reported the results of a study of Vioxx and
Celebrex clinical trials. The JAMA article asserted that available
data raised a “cautionary flag” about the risk of CV events
associated with COX-2 inhibitors. App. at 748. It also stated that
“[c]urrent data would suggest that use of selective COX-2
inhibitors might lead to increased cardiovascular events.” App. at
752. The day before that article was published, Bloomberg News
reported the statement of a Merck scientist that “[w]e already have
8
additional data beyond what they cite, and the findings are very,
very reassuring. VIOXX does not result in any increase in
cardiovascular events compared to placebo.” App. at 539. The
JAMA article garnered extensive coverage. Some securities
analysts responding to the article on the date of its publication
referred to the basic content of the article as “not new news,” App.
at 2749, and noted that the FDA “debated many of the same issues
in February of this year,” at the AAC panel hearing. App. at 2751.
The day after the JAMA article’s publication, Merck issued
a press release stating that it “stands behind the overall and
cardiovascular safety profile . . . of VIOXX.” App. at 540. Merck
also sent “‘Dear Doctor’ letters to physicians throughout the
country disparaging the article as ‘not based on any new clinical
study’ and assuring the physicians that Merck ‘stands behind the
overall and cardiovascular safety profile’ of VIOXX.’” App. at
540.
E. FDA Warning Letter (September 21, 2001)
On September 21, 2001, the FDA posted on its website a
warning letter that its Division of Drug Marketing, Advertising,
and Communications (“DDMAC”) had sent to Merck four days
earlier regarding its marketing and promotion of Vioxx. In the
letter, the DDMAC stated that Merck’s “promotional activities and
materials” for the marketing of Vioxx were “false, lacking in fair
balance, or otherwise misleading in violation of the Federal Food,
Drug, and Cosmetic Act (the Act) and applicable regulations.”
App. at 713. The letter explained:
You have engaged in a promotional campaign for Vioxx
that minimizes the potentially serious cardiovascular
findings that were observed in the [VIGOR] study, and thus,
misrepresents the safety profile for Vioxx. Specifically,
your promotional campaign discounts the fact that in the
VIGOR study, patients on Vioxx were observed to have a
four to five fold increase in myocardial infarctions (MIs)
compared to patients on the comparator non-steroidal anti-
inflammatory drug (NSAID), Naprosyn (naproxen).
9
Although the exact reason for the increased rate of MIs
observed in the Vioxx treatment group is unknown, your
promotional campaign selectively presents the following
hypothetical explanation for the observed increase in MIs.
You assert that Vioxx does not increase the risk of MIs and
that the VIGOR finding is consistent with naproxen’s ability
to block platelet aggregation like aspirin. That is a possible
explanation, but you fail to disclose that your explanation is
hypothetical, has not been demonstrated by substantial
evidence, and that there is another reasonable explanation,
that Vioxx may have pro-thrombotic properties.
App. at 713. The letter also directed Merck to issue “Dear
Healthcare provider” letters “to correct false or misleading
impressions and information.” App. at 719.
The FDA warning letter received widespread coverage by
the media and securities analysts. Although many media reports
focused on the mere fact of the warning letter,4 securities analysts
4
A few representative examples follow: Reuters -- “U.S.
regulators have charged . . . Merck . . . with misleading doctors
about its blockbuster painkiller Vioxx with promotions that
downplayed a possible risk of heart attacks.” App. at 2353.
Associated Press -- “Merck has argued that [the VIGOR study
results make] Vioxx falsely look[] risky because naproxen thins the
blood . . . and thus protect[s] against heart attacks. . . . ‘In fact, the
situation is not at all clear,’ [according to] the FDA . . . .” App. at
2360. USA Today -- “Merck’s marketing efforts, aimed mainly at
doctors, have minimized Vioxx’s known and potential
cardiovascular risks, the FDA wrote in an eight-page ‘warning
letter’ . . . . So far this year, the FDA has sent drug companies
fewer than a dozen warning letters, which the agency reserves for
activities that raise significant public health concerns.” App. at
2355. Wall Street Journal -- “Federal regulators warned Merck &
Co. for improper marketing of its blockbuster arthritis drug Vioxx,
saying the company had misrepresented the drug’s safety profile
and minimized its potential risks. . . . While the FDA sends out
dozens of routine citations annually, it issues only a handful of
10
tended to emphasize the impact the warning letter would likely
have on the prospective Vioxx labeling changes (which were not
forthcoming until April 2002),5 Merck’s ongoing promotional
efforts,6 and Merck’s position in the market.7 A report issued by
UBS Warburg explained, “[t]he FDA pointed out that there is no
definitive study proving or disproving either conclusion [regarding
the higher incidence of CV events associated with Vioxx in the
VIGOR study]. . . . The FDA’s position appears similar to our own,
which is that the data available to date are simply not definitive.”
App. at 2768. Nonetheless, securities analysts were of one voice
in their projections for Merck and Vioxx; analysts from CIBC
World Markets, Credit Suisse First Boston (“CSFB”), Dain
Rauscher, Lehman Brothers, UBS Warburg, SG Cowen, and
these more-serious warning letters each year.” App. at 2361.
5
For example, a report issued by Lehman Brothers stated:
“We do not believe this letter will be predictive of the FDA’s
actions on the pending Vioxx label change. . . . Warning letters of
this nature are certainly not unusual and in fact almost a staple of
the pharmaceutical industry today. . . . As pointed out in the FDA
warning letter, DDMAC does not dispute Merck’s claims.” App.
at 2765-66.
6
One report issued by Merrill Lynch stated: “The FDA
issued a warning letter to Merck . . . [and] is looking for Merck to
cease all violative promotional activities . . . . We do not see how
this issue can be helpful to Merck in promoting Vioxx.” App. at
2752.
7
A Dain Rauscher report focused on Vioxx’s position in the
actual marketplace, i.e., the doctor’s office: “We believe th[e FDA
warning letter] is unlikely to significantly alter physicians’
prescribing practices [because it] is likely that these issues are
already common knowledge in the medical community . . . .” App.
at 2762. Meanwhile, a CIBC World Markets report considered
how the warning letter might impact Merck’s stock price: “The
FDA warning letter as well as a recent JAMA article raising
concerns of cardiovascular risk will continue to pressure the stock,
now trading close to its 52-week low.” App. at 2755.
11
Morgan Stanley all maintained their ratings for Merck stock at
“buy” or “hold” and/or continued to project increased future
revenues for Vioxx.
In the five days between September 20, 2001 and September
25, 2001, Merck’s stock price declined by $4.16, or 6.6%, closing
at $59.11 on September 25. Reuters reported this drop on
September 25, explaining that “[s]hares of Merck & Co. fell . . .
after U.S. regulators accused the firm of making unsubstantiated
claims about its hot-selling arthritis drug Vioxx and downplaying
a possible risk of heart attack from taking the medicine.” App. at
2357. By October 1, 2001, however, Merck’s stock price had
rebounded to $64.66, $1.39 higher than its closing price before the
warning letter was made public just over a week earlier.
F. Additional Vioxx Lawsuits (September 27, 2001)
A consumer fraud lawsuit was filed against Merck on behalf
of Vioxx users on September 27, 2001. A second product liability
lawsuit and a personal injury lawsuit followed shortly thereafter.
In articulating their allegations of fraud and misrepresentations by
Merck to consumers and Vioxx users, the consumer fraud and
product liability suits relied in large part on the JAMA article, the
FDA warning letter, and various media reports concerning Vioxx.
G. New York Times Article (October 9, 2001)
On October 9, 2001, the New York Times published an
article about COX-2 inhibitors entitled “The Doctor’s World; For
Pain Reliever, Questions of Risk Remain Unresolved.” App. at
653. The article reported on “troubling questions about whether
Vioxx may have an unexpected side effect -- a very slight increase
in the risk of heart attack.” App. at 653. However, the article
explained that “[t]he risk is hypothesized, not proved,” and that
“leading arthritis specialists . . . say that they are not concerned and
that they prescribe the drugs for patients who may have heart
disease.” App. at 653. The article noted that “[a]t issue is the
subtle question of what counts as evidence,” App. at 653,
explaining that the risk that COX-2 inhibitors cause blood-clotting
was originally posed as a theory a few years earlier by a scientist
12
from the University of Pennsylvania.
The article addressed defendant Scolnick’s statements at
length. According to the article, Scolnick said that Merck
“look[ed] specifically for excess heart attacks and strokes in” the
VIGOR study and found a higher incidence in the patients taking
Vioxx. App. at 654. “‘There are two possible interpretations,’ Dr.
Scolnick said. ‘Naproxen lowers the heart attack rate, or Vioxx
raises it.’” App. at 654. The article went on, “while [Merck]
announced the heart attack findings to doctors and the public, it
looked back at its data from studies using different drugs or dummy
pills in comparison to Vioxx. It found no evidence that Vioxx
increased the risk of heart attacks, Dr. Scolnick said.” App. at 654.
“He said that the company decided that ‘the likeliest interpretation
of the data is that naproxen lowered . . . the thrombotic event rate’
. . . . He added that without the theoretical question raised by [the
University of Pennsylvania scientist], ‘no one would have a
question remaining in their mind that their [sic] might be an
additional interpretation.’” App. at 654. The article reported
Scolnick as conceding that “none of the findings to date are enough
to prove that the issue is fully resolved. That lack of proof is why
the F.D.A. demanded that Merck explain both sides of the
hypothesis, telling doctors and patients that it is not known whether
naproxen protects against heart attacks or Vioxx makes them more
likely.” App. at 654.
There was no significant movement in Merck’s stock price
following the publication of the New York Times article.
H. Vioxx’s Labeling Modified to Include CV Risks
(April 2002)
Merck was not required to include the risk of CV events in
its labeling until April 2002. The labeling ultimately incorporating
that information explained the VIGOR study results and stated,
“the risk of developing a serious cardiovascular thrombotic event
was significantly higher in patients treated with VIOXX . . . as
compared to patients treated with naproxen . . . . The significance
of the cardiovascular findings . . . is unknown.” App. at 553. This
language was incorporated into the “precautions” section of the
13
Vioxx labeling, rather than the “warnings” section. In a conference
call discussing the labeling changes, a Merck spokesperson
reiterated the company’s “belief that the effect seen in VIGOR
were [sic] the results of the anti-platelet effect of naproxen. . . . So,
I think that’s a position Merck has always had and now its [sic]
quite clearly laid out in the labeling.” App. at 559.
I. Falling Vioxx Sales and the Harvard Study (October
2003)
On October 22, 2003, Reuters published an article entitled
“Merck to Cut 4,400 Jobs, posts Flat Earnings,” in which it
reported that Merck was “hurt by falling sales of arthritis medicine
VIOXX and a paucity of profitable new drugs. . . . The arthritis
drug is suffering from clinical trial data suggesting it might slightly
raise the risk of heart attacks . . . .” App. at 570. That day,
Merck’s stock price dropped from $48.91 to $45.72, down 6.5%.
On October 30, 2003, the Wall Street Journal published an
article entitled “VIOXX Study Sees Heart-Attack Risk,” which
addressed a recent study by the Harvard-affiliated Brigham and
Women’s Hospital in Boston that found an increased risk of heart
attack in patients taking Vioxx compared with patients taking
Celebrex and placebo (the “Harvard study”). App. at 571.
According to the article, “[i]n the first 30 days, the researchers
found, VIOXX was linked to a 39% increased heart-attack risk
compared with Celebrex. Between 30 and 90 days, that increased
relative risk was 37%.” App. at 571. A researcher stated that this
was “the best study to date” and that it “greatly substantiates our
concerns about the cardiac side effects” of Vioxx. App. at 571.
Merck’s stock price dropped below the S&P 500 Index
during this time, and did not rise above that index during the
remainder of the class period.
J. Merck Withdraws Vioxx From the Market
(September 2004)
On September 30, 2004, Merck announced that it was
withdrawing Vioxx from the market based on a new study showing
14
an “increased risk of confirmed cardiovascular events beginning
after 18 months of continuous therapy.” App. at 584. Merck’s
stock price dropped more than $12 per share that day, to close at
$33.00, down 27% from the previous day’s close. Securities
analysts expressed their surprise at the suddenness of Merck’s
action.
On November 1, 2004, the Wall Street Journal reported,
“internal Merck e-mails and marketing materials as well as
interviews with outside scientists show that the company fought
forcefully for years to keep safety concerns from destroying the
drug’s commercial prospects.” App. at 589. Merck’s stock price
dropped another 9.7% based on this news. The news, which was
first published nearly a year after Appellants filed their complaint,
prompted one securities analyst to remark, “new information
indicates to us that the situation might not be as innocent as we
thought. . . . We recommend that investors sell Merck shares.”
App. at 594.
II.
Procedural History
The first class action securities complaint initiating this
lawsuit was filed on November 6, 2003, just weeks after the media
reported the results of the Harvard study and declining Vioxx sales.
After numerous nationwide class actions were consolidated,
Appellants filed a fourth amended consolidated class action
complaint. The complaint alleged that “Defendants’ statements
and omissions during the Class Period materially misrepresented
the safety and commercial viability of VIOXX,” App. at 489, in
violation of sections 11, 12(a)(2), and 15 of the Securities Act of
1933, sections 10(b), 20(a), and 20A of the Securities Exchange
Act of 1934, and Rule 10b-5 promulgated thereunder.
Merck moved to dismiss Appellants’ claims on the grounds
that they were time-barred and that Appellants had failed to state
a claim. The District Court granted that motion on the basis that
the claims were time-barred. In re Merck & Co., Inc. Sec.,
Derivative & “ERISA” Litig., 483 F. Supp. 2d 407, 425 (D.N.J.
15
2007).8 Appellants timely filed a notice of appeal.
III.
Jurisdiction and Standard of Review
The District Court had jurisdiction over this action pursuant
to section 22 of the Securities Act, 15 U.S.C. § 77v; section 27 of
the Securities Exchange Act, 15 U.S.C. § 78aa; and 28 U.S.C. §
1331. We have jurisdiction pursuant to 28 U.S.C. § 1291. We
exercise plenary review over the District Court’s dismissal of
Appellants’ claims for failure to comply with the statute of
limitations. DeBenedictis v. Merrill Lynch & Co., Inc., 492 F.3d
209, 215 (3d Cir. 2007). Because the District Court granted
Merck’s motion to dismiss, “[w]e must ‘accept as true all
allegations in the complaint and all reasonable inferences that can
be drawn therefrom, and view them in the light most favorable to
the non-moving party.’” Id. (quoting Rocks v. City of
Philadelphia, 868 F.2d 644, 645 (3d Cir. 1989)). The dismissal
must be upheld only “‘if it appears to a certainty that no relief
could be granted under any set of facts which could be proved.’”
Id. (quoting D.P. Enters., Inc. v. Bucks County Cmty. Coll., 725
F.2d 943, 944 (3d Cir. 1984)).
IV.
Discussion
The relevant statutes each contain their own statute of
limitations. A complaint alleging “fraud, deceit, manipulation, or
contrivance” under the Securities Exchange Act “may be brought
not later than the earlier of . . . 2 years after the discovery of the
facts constituting the violation; or . . . 5 years after such violation.”
8
The District Court did not address Merck’s argument that
the allegations contained in the fourth amended consolidated class
action complaint failed to satisfy the heightened standards of the
Private Securities Litigation Reform Act of 1995 for pleading
scienter, and we do not express any opinion on this issue.
16
28 U.S.C. § 1658(b). Claims under the Securities Act are subject
to a shorter, one-year limitation period from the time of discovery,
but in no event may be filed later than three years after the public
offering or sale of the security. 15 U.S.C. § 77m. Thus, if
Appellants knew of the basis for their claims prior to November 6,
2001, two years before the first securities complaint was filed, all
of their claims are barred by the statute of limitations. See
DeBenedictis, 492 F.3d at 216.
“Whether the plaintiffs, in the exercise of reasonable
diligence, should have known of the basis for their claims depends
on whether they had ‘sufficient information of possible
wrongdoing to place them on ‘inquiry notice’ or to excite ‘storm
warnings’ of culpable activity.’” Benak ex rel. Alliance Premier
Growth Fund v. Alliance Capital Mgmt., L.P., 435 F.3d 396, 400
(3d Cir. 2006) (quoting In re NAHC, Inc. Sec. Litig., 306 F.3d
1314, 1325 (3d Cir. 2002)). This is an objective question; thus, an
investor is not on inquiry notice until a “‘reasonable investor of
ordinary intelligence would have discovered the information and
recognized it as a storm warning.’” In re NAHC, 306 F.3d at 1325(quoting Mathews v. Kidder, Peabody & Co., 260 F.3d 239, 252
(3d Cir. 2001)).
“If the existence of storm warnings is adequately established
the burden shifts to the plaintiffs to show that they exercised
reasonable due diligence and yet were unable to discover their
injuries.” DeBenedictis, 492 F.3d at 216 (citations, alterations, and
internal quotation marks omitted). Here, the District Court held
that Appellants were on inquiry notice of their claims no later than
October 9, 2001, the date the New York Times published the article
reporting that defendant Scolnick “acknowledged that Merck knew
that the cardioprotective effect of naproxen was not proven and,
further, that Merck admitted that VIOXX may raise the risk of
heart attack or other thrombotic event.” In re Merck, 483 F. Supp.
2d at 419. The Court also noted what it characterized as the
“overwhelming collection of information signaling deceit by Merck
with respect to the safety of VIOXX [that] had accumulated in the
public realm” by that date, in particular, the FDA warning letter.
Id. In concluding that sufficient storm warnings of fraud existed
more than two years prior to the filing of Appellants’ complaint,
17
the District Court observed that Appellants’ “position that their
claims did not accrue until the existence of fraud was a probability,
as opposed to a possibility . . . is simply not supported by Third
Circuit law.” Id. at 422. Finally, noting that Appellants had “not
argued that they conducted a diligent investigation, and nothing in
the Complaint demonstrates that they were unable to uncover
pertinent information during the limitations period,” the Court
concluded that Appellants’ claims were time-barred and granted
Merck’s motion to dismiss. Id. at 424.
A. Principles of Inquiry Notice
Before reviewing the District Court’s decision, we must
address an ambiguity in our inquiry notice jurisprudence.
Appellants contend that the statute of limitations does not begin to
run until there is sufficient evidence of probable, rather than
possible, wrongdoing by the defendants. Predictably, Merck
supports the latter standard, arguing that inquiry notice may be
triggered by evidence of possible wrongdoing. Both formulations
find support in this court’s precedents. Compare DeBenedictis,
492 F.3d at 216 (Inquiry notice may be established by proof of
“‘financial, legal, or other data that would alert a reasonable person
to the probability that misleading statements or significant
omissions had been made.’”) (quoting In re NAHC, 306 F.3d at1326-27 n.5) (emphasis added), with Benak, 435 F.3d at 400(“‘Whether the plaintiffs . . . should have known of the basis for
their claims depends on whether they had “sufficient information
of possible wrongdoing to place them on ‘inquiry notice’ . . . .”’”)
(quoting In re NAHC, 306 F.3d at 1325) (emphasis added). We
therefore take this opportunity to clarify the standard for inquiry
notice in this circuit.
Our first comprehensive discussion of the appropriate
standard for inquiry notice took place in the context of a claim filed
pursuant to the Racketeer Influenced and Corrupt Organizations
Act (“RICO”).9 See Mathews, 260 F.3d at 241. In Mathews,
9
Quoting extensively from Mathews, we recently reiterated
the inquiry notice standard for RICO claims in Cetel v. Kirwan
18
investors in low-risk securities sued their broker after the securities
had lost more than half their value, alleging that the broker misled
them about the nature of the funds and charged excessive fees and
commissions. We affirmed the district court’s grant of summary
judgment for the broker because the complaint was time-barred.
Id. at 244. In analyzing whether plaintiffs’ suit was filed before
RICO’s statute of limitations had run, we applied a two-pronged
test derived from the inquiry notice standard other courts had
applied in the context of securities fraud claims. Id. at 251-52.
First, we noted the requirement to make an objective inquiry
into whether the defendant had met its burden “to show the
existence of ‘storm warnings.’” Id. at 252. We explained that
storm warnings “may take numerous forms,” such as “‘any
financial, legal or other data that would alert a reasonable person
to the probability that misleading statements or significant
omissions had been made.’” Id. (quoting unpublished district court
opinion).10 Second, we described an inquiry, “both subjective and
objective,” into whether the plaintiffs had met their burden “to
show that they exercised reasonable due diligence and yet were
unable to discover their injuries.” Id. We then noted our
agreement with the Court of Appeals for the Seventh Circuit that
courts should be “mindful of the dangers in adopting too broad an
interpretation of inquiry notice.” Id. at 253 (citing Law v. Medco
Research, Inc. (“Medco II”), 113 F.3d 781, 786 (7th Cir. 1997);
Fujisawa Pharm. Co. v. Kapoor, 115 F.3d 1332, 1335 (7th Cir.
1997)).
A year later, we applied this standard to claims pleaded
Financial Group, Inc., 460 F.3d 494, 506-07 (3d Cir. 2006).
10
Immediately before using this language regarding a
“probability” of fraud, we noted without criticism that the district
court had also framed the first prong of the inquiry notice standard
as “‘whether the plaintiffs knew or should have known of the
possibility of fraud (“storm warnings”) . . . .’” Mathews, 260 F.3d
at 251-52 (quoting unpublished district court opinion). Thus, there
is no basis to conclude that we rejected the notion of a possibility
standard at that time.
19
under the federal securities laws. See In re NAHC, 306 F.3d at
1318. The shareholders’ claims in that case arose from a health
care provider’s collapse after the federal government enacted
regulations that negatively impacted the provider’s long-term care
services business. Id. at 1318-21. In formally adopting the inquiry
notice standard for securities claims, we stated that “[w]hether the
plaintiffs, in the exercise of reasonable diligence, should have
known of the basis for their claims depends on whether they had
‘sufficient information of possible wrongdoing to place them on
“inquiry notice” or to excite “storm warnings” of culpable
activity.’” 11 Id. at 1325 (quoting Gruber v. Price Waterhouse, 697
F. Supp. 859, 864 (E.D. Pa. 1988)). We explained that “[p]laintiffs
need not know all of the details or ‘narrow aspects’ of the alleged
fraud to trigger the limitations period; instead, the period begins to
run from ‘the time at which plaintiff should have discovered the
general fraudulent scheme.’” Id. at 1326 (quoting In re Prudential
Ins. Co. Sales Practices Litig., 975 F. Supp. 584, 599 (D.N.J.
1997)).
In affirming the district court’s dismissal of the plaintiffs’
claim arising from the impact of the federal regulations on the
defendants’ long-term care services business, we held that a series
of disclosures, which accompanied a drastic decline in the
company’s stock price, id. at 1319, and culminated with the
defendants’ announcement that they were writing off goodwill and
selling their business for nominal consideration, put the plaintiffs
on inquiry notice that previous valuations of goodwill had been
inflated, id. at 1326-27. This holding was bolstered by the
plaintiffs’ admission that the market had written off that business
even before the defendants’ announcement. Id. at 1327.
More recently, in 2006, we considered whether a suit filed
11
In a footnote of the same opinion, however, we stated that
inquiry notice could be established on the basis of “‘data that
would alert a reasonable person to the probability that misleading
statements or significant omissions had been made.’” See In re
NAHC, 306 F.3d at 1325-26 n.5 (quoting Mathews, 260 F.3d at
252).
20
by mutual fund investors against fund advisors who had invested
heavily in Enron was barred by the statute of limitations. Benak,
435 F.3d at 397. In our decision, we dispensed with the probability
language altogether, instead holding that storm warnings could be
triggered by “‘sufficient information of possible wrongdoing . . . .’”
Id. at 400 (quoting In re NAHC, 306 F.3d at 1325). Applying our
inquiry notice standard to the facts of that case, we distinguished
mutual fund investors from direct investors on the ground that
mutual fund investors rely on an intermediary to learn about the
companies in which they have invested. Id. at 401-02.
Nonetheless, because the investors in Benak had access to media
reports about their fund’s large holdings in Enron after that
company went bankrupt, we concluded that the plaintiffs were on
inquiry notice of the fraud by the time the media reported the
bankruptcy. Id. at 402-03.
Finally, in a case decided just last year, we considered
investors’ claims that Merrill Lynch misled them by failing to
disclose that a certain class of mutual fund shares was “never a
rational choice of investment for them and that Merrill brokers
received larger commissions on sales of such shares.”
DeBenedictis, 492 F.3d at 210. Merrill argued that news articles,
National Association of Securities Dealers (“NASD”) press
releases, and the mutual funds’ registration statements put the class
on inquiry notice more than two years before the complaint was
filed and that it should therefore be dismissed as time-barred. Id.
at 214. After quoting the “probability” language first used by the
district court in Mathews, we addressed each category of storm
warnings alleged. We noted that Merrill’s registration statements
disclosed the fee structure for the different classes of shares, which
allowed investors to determine the relative costs and benefits of the
different shares, and the different commissions applying to those
shares. Id. at 216-17. We further concluded that storm warnings
existed because the news reports and press releases identified by
the defendants revealed that many brokers had been disciplined by
the NASD for recommending the very class of shares that
undergirded the plaintiffs’ claims. Id. at 217. Accordingly, we
concluded that the plaintiffs’ claims were time-barred.
As this review of our precedent makes clear, although we
21
have occasionally stated that inquiry notice may be triggered by
evidence alerting an investor to the probability of wrongdoing, we
have just as often emphasized that inquiry notice may be triggered
by sufficient information of possible wrongdoing. This implies
that a probability, in the sense of a nearly certain likelihood, of
wrongdoing is not necessary to trigger storm warnings in this
circuit. Therefore, we reaffirm that “whether the plaintiffs, in the
exercise of reasonable diligence, should have known of the basis
for their claims depends on whether they had sufficient information
of possible wrongdoing to place them on inquiry notice or to excite
storm warnings of culpable activity.” Benak, 435 F.3d at 400(citations, alteration, and internal quotation marks omitted). In so
holding, we note that the majority of courts of appeals to have
addressed the question employ a possibility standard when
evaluating the likelihood of wrongdoing sufficient to constitute
storm warnings. See, e.g., GO Computer, Inc. v. Microsoft Corp.,
508 F.3d 170, 179 (4th Cir. 2007); Tello v. Dean Witter Reynolds,
Inc., 494 F.3d 956, 970 (11th Cir. 2007); Wolinetz v. Berkshire
Life Ins. Co., 361 F.3d 44, 48 (1st Cir. 2004); Ritchey v. Horner,
244 F.3d 635, 639 (8th Cir. 2001); Berry v. Valence Tech., Inc.,
175 F.3d 699, 705 (9th Cir. 1999); Sterlin v. Biomune Sys., 154
F.3d 1191, 1196 (10th Cir. 1998); LaSalle v. Medco Research, Inc.
(“Medco I”), 54 F.3d 443, 444 (7th Cir. 1995); Jensen v. Snellings,
841 F.2d 600, 607 (5th Cir. 1988). But see Newman v. Warnaco
Group, Inc., 335 F.3d 187, 193 (2d Cir. 2003) (“The [existence of]
fraud must be probable, not merely possible.”).
Nonetheless, simply repeating the word “possibility” or
“probability” with ever-increasing frequency and intensity (as both
parties did in their briefs and at oral argument) is hardly useful.
Rather, we review the information set forth by the parties with an
eye toward the practical effect of drawing the inquiry notice line at
a particular date. In this vein, we have emphasized that
“[u]ndergirding the inquiry notice analysis is the assumption that
a plaintiff either was or should have been able, in the exercise of
reasonable diligence, to file an adequately pled securities fraud
complaint as of an earlier date.” Benak, 435 F.3d at 401.
Similarly, the Court of Appeals for the Seventh Circuit, which has
also applied a possibility standard, see Medco I, 54 F.3d at 444, has
reasoned that “[t]he facts constituting [inquiry] notice must be
22
sufficiently probative of fraud–sufficiently advanced beyond the
stage of a mere suspicion, sufficiently confirmed or
substantiated–not only to incite the victim to investigate but also to
enable him to tie up any loose ends and complete the investigation
in time to file a timely suit,” Fujisawa, 115 F.3d at 1335. In other
words, simply stating that a smattering of evidence hinted at the
possibility of some type of fraud does not answer the question
whether there was “sufficient information of possible wrongdoing
. . . to excite storm warnings of culpable activity” under the
securities laws. Benak, 435 F.3d at 400 (citations and internal
quotation marks omitted) (emphasis added).
This concern is reenforced by the heightened pleading
requirements of the Private Securities Litigation Reform Act of
1995 (“PSLRA”), 15 U.S.C. § 78u-4(b).12 Surely, Congress did not
envision a statute of limitations that would open the floodgates to
a rush of premature securities litigation when its primary foray into
this field in recent decades has been to deter poorly pleaded
allegations of securities fraud. See DeBenedictis, 492 F.3d at 217-
18 (noting that “‘the level of particularity in pleading required by
the PSLRA is such that inquiry notice can be established only
where the triggering data “relates directly to the misrepresentations
and omissions” alleged.’”) (quoting Lentell v. Merrill Lynch &
Co., 396 F.3d 161, 171 (2d Cir. 2005)); cf. Mathews, 260 F.3d at
253 (expressing concern about “a flood of untimely litigation” were
we to “adopt[] too broad an interpretation of inquiry notice”).
B. Basis of Appellants’ Claims
Appellants argue that the District Court mischaracterized the
gravamen of their fraud allegations, thereby undermining the
Court’s conclusion that Appellants were on inquiry notice of the
12
The PSLRA requires plaintiffs pleading securities fraud
to “specify each statement alleged to have been misleading, [and]
the reason or reasons why the statement is misleading,” 15 U.S.C.
§ 78u-4(b)(1), and to “state with particularity facts giving rise to a
strong inference that the defendant acted with the required state of
mind,” i.e., scienter, id. § 78u-4(b)(2).
23
alleged wrongdoing. We have repeatedly stated that the
fundamental concern of our analysis is whether plaintiffs were “‘on
inquiry notice of the basis for [their] claims’” prior to the relevant
date triggering the statute of limitations. Benak, 435 F.3d at 400(quoting In re NAHC, 306 F.3d at 1325). Therefore, we must
carefully scrutinize the District Court’s characterization of the basis
for Appellants’ claims and consider how this characterization
affected the Court’s inquiry notice analysis.
First, Appellants contend that the Court mischaracterized the
basis for their claims by focusing on alleged misrepresentations
about Vioxx’s safety profile. In concluding that sufficient storm
warnings existed to put Appellants on inquiry notice, the District
Court considered the “overwhelming collection of information
signaling deceit by Merck with respect to the safety of VIOXX
[that] had accumulated in the public realm” by October 9, 2001. In
re Merck, 483 F. Supp. 2d at 419. Appellants argue that the true
nature of their claims is that Vioxx “was so dangerous that it lacked
any meaningful commercial prospects, or that [Merck’s]
representations in this regard were materially false and misleading
when made . . . .” Appellant’s Br. at 35. The difficulty with this
contention is that Merck’s representations about Vioxx’s
commercial viability are not unrelated to the company’s
representations about the drug’s safety profile. If public
information undermined Merck’s representations about Vioxx’s
safety, a reasonable investor would also likely see such information
as undermining Merck’s representations about Vioxx’s commercial
viability. Indeed, some professional investors connected concerns
about the safety of COX-2 inhibitors to their commercial viability.
Nonetheless, the fact that many securities analysts continued
to maintain strong growth ratings for Vioxx at the same time that
its safety was being questioned is certainly relevant to whether such
questions constituted sufficient information of possible
wrongdoing to trigger storm warnings. Even though there were
analysts who connected Vioxx’s safety to its commercial viability,
it appears that they were not so worried about Vioxx’s safety after
the FDA warning letter was made public that they felt it necessary
to retract their opinions about Vioxx’s future profitability or
24
Merck’s market position.13 In any event, Appellants argue that
even if their claims are properly characterized as alleging
misrepresentations about Vioxx’s safety, the District Court
misinterpreted their claims in another respect.
Appellants contend that their complaint challenges the
veracity of Merck’s statements of opinion and belief regarding the
naproxen hypothesis whereas the District Court analyzed whether
Merck misrepresented the fact that the results of the VIGOR study
could support multiple hypotheses (i.e., that naproxen lowers the
risk of CV events or that Vioxx raises that risk). Thus, they argue
that the District Court mischaracterized their claims by considering
whether there were storm warnings that put them on notice of a
fraud different from that which they have asserted in their
complaint.
We have explained that for “misrepresentations in an
opinion” or belief to be actionable, plaintiffs must show that the
statement was “‘issued without a genuine belief or reasonable
basis’ . . . .” Herskowitz v. Nutri/System, Inc., 857 F.2d 179, 185
(3d Cir. 1988) (quoting Eisenberg v. Gagnon, 766 F.2d 770, 776
(3d Cir. 1985)); accord Va. Bankshares, Inc. v. Sandberg, 501 U.S.
1083, 1095 (1991) (“A statement of belief may be open to
objection . . . as a misstatement of the psychological fact of the
speaker’s belief in what he says.”). Thus, to trigger “storm
warnings of culpable activity,” Benak, 435 F.3d at 400 (citations
and internal quotation marks omitted), in the context of a claim
alleging falsely-held opinions or beliefs, investors must have
13
A September 25, 2001 report by a CSFB analyst illustrates
the interrelatedness of the two propositions: “Recent prescription
trends have indicated that adverse publicity and cardiovascular
concerns have contributed to erosion in Vioxx (as well as
Celebrex) market share within the collective COX-2/NSAID
market.” App. at 2757. On the other hand, the same CSFB report
also “project[ed] Vioxx revenues will increase 42% year over year
to $3.06 billion for 2001, with growth moderating to the +14%
level in 2002 to $3.49 billion. We maintain our Buy rating.” App.
at 2757.
25
sufficient information to suspect that the defendants engaged in
culpable activity, i.e., that they did not hold those opinions or
beliefs in earnest. Appellants’ theory in the complaint is that
Merck’s statements about the validity of the naproxen hypothesis
were falsely-held statements of opinion or belief and that there was
no information available to investors prior to November 6, 2001
that would have led them to suspect that such statements were not
held in earnest. The District Court rejected this argument,
concluding that “[i]t is prepost[e]rous for Plaintiffs to argue that
because they did not have a ‘smoking gun’ that demonstrated that
Defendants’ misrepresentation was even more egregious than the
[FDA] Warning Letter charged, they were not on inquiry notice of
a general fraudulent scheme regarding the safety of VIOXX.” In
re Merck, 483 F. Supp. 2d at 422-23. We disagree.
It is true that “[p]laintiffs cannot avoid the time bar simply
by claiming they lacked knowledge of the details or narrow aspects
of the alleged fraud. Rather, the clock starts when they should
have discovered the general fraudulent scheme.” Benak, 435 F.3d
at 400 (citations and internal quotation marks omitted). The
“fraudulent scheme” referred to must be one “in connection with
the purchase or sale of any security . . . .” 15 U.S.C. § 78j(b).
Appellants have brought a securities fraud action, not a consumer
fraud action, against Merck. See Gavin v. AT & T Corp., 464 F.3d
634, 640 (7th Cir. 2006) (recognizing that securities fraud suits and
consumer fraud suits are not interchangeable); cf. Marine Bank v.
Weaver, 455 U.S. 551, 556 (1982) (“Congress, in enacting the
securities laws, did not intend to provide a broad federal remedy
for all fraud.”). Thus, the fact that the FDA sent a letter to Merck
about its possible misrepresentations in connection with its
promotion of Vioxx to health care professionals would not have
provided a storm warning unless it put Appellants on inquiry notice
of actionable misrepresentations under the securities laws. See
DeBenedictis, 492 F.3d at 218 (finding storm warning where
disclosure was “directly applicable to the representations or
omissions” challenged by plaintiffs). The asserted basis of
Appellants’ claims is that Merck defrauded investors by proposing
and reasserting the naproxen hypothesis at the same time that it
knew the hypothesis was false. We must analyze the existence of
storm warnings relative to that allegation in order to determine
26
whether Appellants were on inquiry notice of the alleged fraud.
C. Existence of Storm Warnings
Because the District Court believed that “[t]he wrongdoing
charged in the [FDA] Warning Letter is . . . the same alleged
misconduct on which the securities fraud claims in this case are
predicated,” the Court asserted that it “might arguably conclude
that the FDA Warning Letter alone excited storm warnings
sufficient to put Plaintiffs on inquiry notice of their claims against
Merck,” but it decided that it “need not make that conclusion,
because the FDA Warning Letter was not issued in a vacuum of
information.” In re Merck, 483 F. Supp. 2d at 419. The Court then
took notice of the JAMA article, the lawsuits filed against Merck
in 2001, and various articles discussing competing explanations for
the results of the VIGOR study. Id. at 419-21. The Court reasoned
that the New York Times article following the FDA warning letter
was especially probative because Scolnick “admitted that Merck
recognized the possibility that VIOXX may increase a user’s risk
of heart attack. It therefore represents a significant departure from
Merck’s company line as to the explanation for the VIGOR study
results.” Id. at 420. The Court then rejected Appellants’ argument
that positive information issued by Merck during this period
dissipated any storm warnings. Id. at 421.
Appellants argue that to the extent the disclosures identified
by Merck might be seen as triggering storm warnings, such storm
warnings were dissipated by Merck’s reassuring statements,14 and
14
We have recognized that “reassurances can dissipate
apparent storm warnings if an investor of ordinary intelligence
would reasonably rely on them to allay the investor’s concerns.”
Benak, 435 F.3d at 402 n.16 (citation, alteration, and internal
quotation marks omitted). “‘Whether reassuring statements justify
reasonable reliance that apparent storm warnings have dissipated
will depend in large part on how significant the company’s
disclosed problems are, how likely they are of a recurring nature,
and how substantial are the “reassuring” steps announced to avoid
their recurrence.’” DeBenedictis, 492 F.3d at 218 (quoting LC
27
are undermined by the failure of the identified disclosures to have
any significant impact on Merck’s stock price or the projections of
securities analysts covering Merck. Merck argues that stock price
movement is irrelevant to the inquiry notice analysis. We cannot
agree. Our past inquiry notice decisions have taken into account
the market reaction to disclosures that purportedly constitute storm
warnings. See, e.g., In re NAHC, 306 F.3d at 1319 (discussing
drastic decline in stock price accompanying disclosures in period
leading up to date of inquiry notice); cf. Benak, 435 F.3d at 403(noting that Enron’s collapse and subsequent bankruptcy triggered
inquiry notice); Mathews, 260 F.3d at 254 (explaining that 30%
drop in funds’ net asset values and 60% decline in distributions
triggered inquiry notice). In Mathews, we explained that “in most
securities fraud actions, the plaintiffs’ [losses] are inextricably
intertwined with the defendant’s misrepresentations. Discovery of
one leads almost immediately to discovery of the other.” Mathews,
260 F.3d at 251. Similarly, in the context of materiality, we have
stated that in “an efficient market, ‘information important to
reasonable investors . . . is immediately incorporated into the stock
price.’” Oran v. Stafford, 226 F.3d 275, 282 (3d Cir. 2000)
(quoting In re Burlington Coat Factory Sec. Litig., 114 F.3d 1410,
1425 (3d Cir. 1997)). “If the disclosure of certain information has
no effect on stock prices, it follows that the information disclosed
was immaterial as a matter of law.” In re NAHC, 306 F.3d at 1330
(citing In re Burlington Coat Factory, 114 F.3d at 1425).
Because information that is material to reasonable investors
is immediately incorporated into the stock price, the effect of a
purported storm warning on the market, while insufficient on its
own to compel the conclusion that inquiry notice has not been
triggered, is, contrary to Merck’s position, relevant to our inquiry.
See Newman, 335 F.3d at 195 (asserting that the court’s “holding
is further supported by the fact that [defendant]’s stock price did
not have any significant movement following” the identified
disclosure); Berry, 175 F.3d at 705 (concluding that the lack of
significant stock movement “bolster[ed]” the conclusion that
Capital Partners, LP v. Frontier Ins. Group, Inc., 318 F.3d 148, 155
(2d Cir. 2003)).
28
inquiry notice had not been triggered).
The District Court (and Merck on this appeal) emphasized
five classes of information, each of which was disclosed on or
before October 9, 2001, which purportedly triggered storm
warnings: (1) articles and reports commenting on the hypothetical
explanations for the results of the VIGOR study; (2) the JAMA
article, which asserted that available data (i.e., VIGOR and a
Celebrex study) raised a “cautionary flag” about the risk of CV
events in COX-2 inhibitors, App. at 748; (3) the FDA warning
letter, which charged Merck with “engag[ing] in a promotional
campaign for Vioxx that minimizes the potentially serious
cardiovascular findings that were observed in the [VIGOR] study,
and thus, misrepresents the safety profile for Vioxx,” App. at 713;
(4) the consumer fraud, product liability, and personal injury
lawsuits filed against Merck throughout 2001; and (5) the New
York Times article, in which Scolnick stated there were “two
possible interpretations” for the VIGOR results, App. at 654.
Because the disclosures in each of these categories ultimately arise
from the results of the VIGOR study, we briefly recap the details
of the study.
VIGOR compared Vioxx to naproxen in the hopes of
establishing that Vioxx had a better GI profile than traditional
NSAIDs. Although those hopes were realized, Merck also learned,
and subsequently notified the public, that “significantly fewer
thromboembolic events were observed in patients taking naproxen”
than patients taking Vioxx. App. at 765. Merck suggested that
naproxen’s effect on platelet aggregation was responsible for this
difference, but conceded that this hypothetical effect “had not been
observed previously in any clinical studies . . . .” App. at 765.
Merck also stated that all other Vioxx trials “showed no indication
of a difference in the incidence of thromboembolic events between
Vioxx, placebo and comparator NSAIDs.” App. at 766.
Securities analysts and the press duly reported the results of
VIGOR and the naproxen hypothesis. For instance, an article
published in Bloomberg News a month after the VIGOR results
were released reiterated Merck’s hypothesis about naproxen’s
effect on platelet aggregation, but noted that “[n]aproxen doesn’t
29
have documented protective effects on the heart,” and quoted an
analyst who stated, “that Vioxx increases cardiac risk . . . may be
true, but it is far too soon to make that kind of judgment.” App. at
2292. Similarly, a J.P. Morgan report from April 2000 noted the
intuitive appeal of the theory that “the thromboembolic event issue
is an ‘NSAID-issue,’” but explained that the “theoretical
cardiovascular protective benefits of Naprosyn . . . have not been
clinically proven . . . .” App. at 2376. In another article, a
spokesperson for the makers of Naprosyn stated, “[t]o our
knowledge, naproxen does not prevent heart attack or stroke . . . .”
App. at 2288. In our view, this category of disclosures does not
constitute storm warnings that Merck misrepresented Vioxx’s
safety profile to investors in a manner that might give rise to a
securities fraud claim. On the contrary, securities analysts and the
press recognized the naproxen hypothesis for what it was, an
unproven hypothesis, and recognized that there was an alternative
hypothesis, “that Vioxx increases cardiac risk . . . .” App. at 2292.
Shortly before the AAC hearing, during which the FDA
considered how Vioxx’s labeling should be modified to incorporate
the results of the VIGOR study, a J.P. Morgan research report
described the effect of NSAIDs such as naproxen on CV events as
“poorly proven” and explained that there was “no way to
retrospectively slice the data to prove the NSAID benefit vs. Vioxx
risk argument . . . .” App. at 2547. At that hearing, defendant
Reicin, the Executive Director of Clinical Research at Merck
Research Laboratories, who argued in support of the naproxen
hypothesis, admitted at the outset that the explanation for the
results of the VIGOR study was uncertain. The first Vioxx product
liability lawsuit (which, incidentally, charged the makers of
Celebrex with identical wrongdoing) followed shortly thereafter,
seeking “additional medical labeling which is presently being
considered by the FDA [in conjunction with the AAC hearing.]”
App. at 1748. Of course, investors, unlike Vioxx patients, were
presumed to be aware of the publicized outcomes of research
studies, such as VIGOR, which underlay the allegations of that
product liability lawsuit. See Benak, 435 F.3d at 401 (explaining
that “a direct investor . . . can be deemed to have consistent
knowledge of his or her securities holdings”).
30
The JAMA article evaluated Vioxx and Celebrex, both
COX-2 selective inhibitors, together; its findings were not limited
to Vioxx. The article concluded, based in part on VIGOR, that
“[c]urrent data would suggest that use of selective COX-2
inhibitors might lead to increased cardiovascular events.” App. at
752. Of course, this is simply the alternative to the naproxen
hypothesis. The JAMA article did not present any data that would
suggest that Merck did not have reason to propose that hypothesis.
Accordingly, it is of little surprise that a Deutsche Banc securities
analyst described the types of questions raised in the JAMA article
as “not new news . . . .” App. at 2749. Moreover, Merck issued
reassuring statements the day before and the day after the article
was published. Again, we are of the view that the JAMA article,
taken on its own, did not constitute sufficient information of
possible wrongdoing under the securities laws so as to raise a storm
warning of culpable activity under the securities laws.
The FDA warning letter demands more scrutiny. In
analyzing the effect of that letter through the prism of inquiry
notice, we must not lose focus of the nature of the allegations in the
letter and the scope of the FDA’s regulatory authority. The FDA
targeted Merck’s “promotional campaign for Vioxx,” App. at 713,
under its authority to regulate prescription drug advertisements, see
21 U.S.C. § 352(n); see generally Pa. Employees Benefit Trust
Fund v. Zeneca Inc., 499 F.3d 239, 248-49 (3d Cir. 2007)
(discussing the FDA’s authority over prescription drug
advertising). The letter focused on three distinct components of the
promotional campaign that the FDA found of concern: (1) six
promotional audio conferences, presumably aimed at health care
professionals such as doctors and pharmacists; (2) a press release
dated May 22, 2001 entitled “Merck Confirms Favorable
Cardiovascular Safety Profile of Vioxx,” App. at 718; and (3) oral
representations made by Merck sales representatives, again,
presumably to health care professionals. The FDA chastised
Merck’s promotional campaign for “discount[ing] the fact that in
the VIGOR study, patients on Vioxx were observed to have a four
to five fold increase in myocardial infarctions (MIs) compared to
patients on” naproxen, and “selectively present[ing]” the naproxen
hypothesis as the reason for the incidence of increased CV events.
App. at 713. The FDA stated that Merck’s promotional campaign
31
“fail[ed] to disclose that [its] explanation is hypothetical, has not
been demonstrated by substantial evidence, and that there is
another reasonable explanation, that Vioxx may have pro-
thrombotic properties.” App. at 713. For a number of reasons, we
are hesitant to conclude that the FDA warning letter was sufficient
to trigger inquiry notice.
To begin with, the FDA was acting as a regulator of drug
advertising, rather than as a regulator of the securities markets.
Thus, contrary to Merck’s contention at oral argument, the FDA’s
actions are hardly analogous to allegations of accounting fraud
issued by the SEC, which regulates the securities markets. Indeed,
the FDA’s drug advertising regulations and the securities laws
provide wholly different standards with respect to what constitutes
a misrepresentation. FDA regulations provide that advertisements
must not be “lacking in fair balance,” 21 C.F.R. § 202.1(e)(6), and
prohibit advertisements that “[c]ontain[] a representation or
suggestion that a drug is safer than it has been demonstrated to be
by substantial evidence or substantial clinical experience . . . or
otherwise selects information from any source in a way that makes
a drug appear to be safer than has been demonstrated,” id. §
202.1(e)(6)(iv). In contrast, under the securities laws, “a fact or
omission is material only if ‘there is a substantial likelihood that it
would have been viewed by the reasonable investor as having
significantly altered the “total mix” of information’ available to the
investor.” In re NAHC, 306 F.3d at 1330 (quoting Basic Inc. v.
Levinson, 485 U.S. 224, 231-32 (1988)).
Second, the FDA’s description of the truth about the
VIGOR study is quite similar to the evidence that Merck had long
acknowledged and which the market had incorporated.
Specifically, the FDA stated that the naproxen hypothesis “is
hypothetical, has not been demonstrated by substantial evidence,
and that there is another reasonable explanation, that Vioxx may
have pro-thrombotic properties.” App. at 713. This information is
implicit in Merck’s long-standing admission that the posited anti-
coagulant effect of naproxen “on [CV] events had not been
observed previously in any clinical studies for naproxen.” App. at
765. On the basis of Merck’s public announcements, securities
analysts discussed the existence of “a ‘Vioxx risk’ hypothesis” over
32
seven months before the FDA warning letter was issued. App. at
2547. Indeed, the FDA did not charge that the naproxen hypothesis
was wrong or that Merck did not believe in the validity of its
hypothesis; rather, the agency simply directed Merck to be more
clear about the widely known alternative hypothesis in its dealings
with health care professionals and, presumably, consumers.
Third, two of the three components of the promotional
campaign subject to the FDA’s reprimand consisted of statements
made to health care professionals in the course of targeted audio
conferences and personal conversations. The third component of
the promotional campaign targeted by the FDA was the press
release, but that press release merely repeated the same information
that was first contained in the VIGOR press release, i.e.,
“significantly fewer heart attacks were observed in patients taking
naproxen . . . compared to the group taking Vioxx,” “the VIGOR
finding is consistent with naproxen’s ability to block platelet
aggregation by inhibiting COX-1,” “[t]his is the first time this
effect of naproxen on cardiovascular events has been observed in
a clinical study,” and “[o]ther potential explanations” for the results
were possible. Press Release, Merck & Co., Inc., Merck Confirms
Favorable Cardiovascular Safety Profile of Vioxx(R) (May 22,
2001) (available on PR Newswire and LexisNexis).
Finally, we consider the effect the FDA warning letter had
on the market. Merck’s stock price dipped slightly following the
disclosure of the FDA warning letter before closing higher than it
did before that disclosure just a week and a half later. Although the
lack of significant movement in Merck’s stock price following the
FDA warning letter is not conclusive, it supports a conclusion that
the letter did not constitute a sufficient suggestion of securities
fraud to trigger a storm warning of culpable activity under the
securities laws. See, e.g., Berry, 175 F.3d at 705 (asserting that the
“negligible impact” of an alleged storm warning on defendant’s
stock price bolstered conclusion that inquiry notice was not
triggered). This conclusion is also supported by the fact that more
than a half-dozen securities analysts continued to maintain their
ratings for Merck stock and/or project increased future revenues for
Vioxx after the warning letter was made public.
33
Merck also emphasizes the three additional lawsuits filed
after the FDA warning letter. Of course, none of these lawsuits
alleged securities fraud. Rather, they alleged consumer fraud,
product liability, and personal injury claims. The claims in those
lawsuits alleged that Merck failed to provide publicly available
information to Vioxx consumers, rather than to Merck investors.
Cf. In re Ames Dep’t Stores, Inc. Note Litig., 991 F.2d 968, 980
(2d Cir. 1993) (stating that the different concerns of debt and
equity holders may call for distinct inquiry notice dates for the two
classes of investors).
Finally, we question the District Court’s conclusion that the
New York Times article constituted a storm warning. The District
Court reasoned that defendant Scolnick’s statements in that article
constituted “a significant departure from Merck’s company line as
to the explanation for the VIGOR study results.” In re Merck, 483
F. Supp. 2d at 420. But Scolnick did not abandon the naproxen
hypothesis; rather, he reiterated that Merck “found no evidence that
Vioxx increased the risk of heart attacks” when it looked back at
its data comparing Vioxx to other drugs and placebos and “that ‘the
likeliest interpretation of the data is that naproxen lowered . . . the
thrombotic event rate’ . . . .” App. at 654. 15 Even in the wake of
the FDA warning letter, then, Merck continued to reassure the
investing public that Merck stood behind the naproxen hypothesis,
while acknowledging that another explanation (i.e., that Vioxx
causes CV events) remained a possibility. See Benak, 435 F.3d at402 n.16 (“Reassurances can dissipate apparent storm warnings if
an investor of ordinary intelligence would reasonably rely on them
to allay the investor’s concerns.”) (citation and internal quotation
marks omitted). It is also notable there was no “significant
movement” of Merck’s stock price following the article’s
publication. Newman, 335 F.3d at 195. Thus, we cannot conclude
as a matter of law that this article constituted a storm warning.
15
The New York Times article also explained that “[t]he
risk is hypothesized, not proved,” and that “leading arthritis
specialists . . . say that they are not concerned and that they
prescribe the drugs for patients who may have heart disease.” App.
at 653.
34
In summary, we conclude that the District Court acted
prematurely in finding as a matter of law that Appellants were on
inquiry notice of the alleged fraud before October 9, 2001. As of
that date, market analysts, scientists, the press, and even the FDA
agreed that the naproxen hypothesis was plausible, at the very least.
None suggested that Merck believed otherwise. Accordingly, in
April 2002, the FDA approved a labeling change for Vioxx which
stated that “[t]he significance of the cardiovascular findings [from
the VIGOR study] is unknown.” App. at 553. Merck continued to
reassure the investing public at this time, explaining that the
naproxen hypothesis was “a position Merck has always had and
now its [sic] quite clearly laid out in the labeling.” App. at 559.
On the record before us, there is no reason to suspect that Merck
did not believe the naproxen hypothesis until the Harvard study in
2003 revealed an increased risk of heart attack in patients taking
Vioxx compared with patients taking Celebrex and placebo. This
study for the first time belied Merck’s repeated assurances that
naproxen was responsible for the disparity in CV events in VIGOR
and that Vioxx did not have a higher incidence of CVs compared
to placebo or comparator NSAIDs, such as Celebrex.16
16
There are two statements in the dissent, although arguably
going to minor issues, that call for a response. The dissent states
that Scolnick’s statement quoted in the October 9, 2001 New York
Times article was “the first time [the statement that the VIGOR
results could be explained by either the effect of naproxen or
Vioxx] had been made by the company.” Dissent Typescript op. at
46-47. In fact, as noted above, Alise Reicin, the Executive
Director of Clinical Research at Merck Research Laboratories, had
testified as to that possibility at the FDA’s hearing before the AAC
as early as February 8, 2000, more than eight months before the
New York Times article. See supra p. 7.
Second, to the extent that the dissent suggests that the
majority holds that fluctuations in stock price and analysts’ ratings
and projections are necessary to a finding of storm warnings,
Dissent Typescript op. at 47, a rereading of the majority opinion
will make clear that the majority agrees with the dissent that such
factors are relevant to the storm warnings inquiry, but not required.
35
V.
Conclusion 17
For the reasons set forth, we will reverse the judgment of
dismissal and remand to the District Court for further proceedings
consistent with this opinion.
See supra p. 28.
It is ironic that the dissent, although noting what might be
viewed as Merck’s misrepresentations, would apply the statute of
limitations to deprive plaintiffs of the opportunity to prove a viable
case against Merck for such misrepresentations.
17
Because we have concluded that the District Court erred
in finding Appellants on inquiry notice of the alleged fraud at this
stage of the litigation, we do not address Appellants’ remaining
arguments regarding the claims of plaintiffs who purchased stock
after October 9, 2001 and the viability of Appellants’ section 20A
claims.
36
In re: Merck & Co., et al.
Nos. 07-2431/2432
ROTH, Circuit Judge, dissenting.
I believe “storm warnings” alerting a reasonable investor
of possible culpable activity on the part of Merck were evident
more than two years prior to the filing of appellants’ complaint.
In particular, I believe that the FDA’s September 17, 2001,
warning letter, in and of itself, provided sufficient “storm
warnings” to put the appellants on inquiry notice of their claims
regardless of any significant change in stock price or analysts’
stock ratings or projections at that time. I therefore respectfully
dissent.
Under the “inquiry notice” test, the statute of limitations
for securities claims “begins to run when the plaintiffs
‘discovered or in the exercise of reasonable diligence should
have discovered the basis for their claim’ against the defendant.”
Benak v. Alliance Capital Management L.P., 435 F.3d 396, 400
(3d Cir. 2006) (quoting In re NAHC, Inc. Securities Litigation,
306 F.3d 1314, 1325 (3d Cir. 2002) (citations omitted)). In
order to establish that plaintiffs were on inquiry notice, a
defendant must demonstrate that, as of a particular date, there
existed “storm warnings” sufficient to alert “a reasonable
investor of ordinary intelligence” to “possible wrongdoing” on
the part of defendants. Id. (quoting In re NAHC, 306 F.3d at
1325) (explaining that the question is whether plaintiffs had
37
“sufficient information of possible wrongdoing to place them on
‘inquiry notice’ or to excite ‘storm warnings’ of culpable
activity”) (emphasis added).
Furthermore, it is well established that “[t]he existence of
storm warnings is a totally objective inquiry[,]” that is based on
whether a “reasonable investor of ordinary intelligence would
have discovered the information and recognized it as a storm
warning[,]” Mathews v. Kidder Peabody & Co., Inc., 260 F.3d
239, 252 (3d Cir. 2001) (emphasis added); see also In re NAHC,
306 F.3d at 1325. We do not require that plaintiffs “know all of
the details or ‘narrow aspects’ of the alleged fraud to trigger the
limitations period[,]” but rather “the period begins to run from
the time at which plaintiff should have discovered the general
fraudulent scheme.” In re NAHC, 306 F.3d at 1326 (internal
quotations and citations omitted). Most importantly, we
recognize that triggering data for “storm warnings” may include
any information that would alert a reasonable investor to the
possibility that the defendants engaged in the “general
fraudulent scheme” alleged in the complaint. Id. (emphasis
added). Finally, such triggering data must “relate[] directly to
the misrepresentations and omissions alleged.” DeBenedictis,
492 F.3d at 217-18 (quoting Lentell v. Merrill Lynch & Co.,
Inc., 396 F.3d 161, 171 (2d Cir. 2005)).
In applying the above inquiry notice standard to the
instant case, I am reminded of a classic fairytale: The
Emperor’s New Clothes, by Danish author and poet, Hans
38
Christian Anderson.18 As the child in The Emperor’s New
Clothes saw – that the Emperor walked naked down the street –
any reasonable investor reading the FDA’s September 17, 2001,
warning letter could see the problem with Vioxx – the
misrepresentation of its safety profile and the “possibility” that
Merck had fraudulently misrepresented the cardiovascular safety
of its “blockbuster” product. The warning letter to Merck,
which was published on the FDA’s public website, stated in
pertinent part:
You have engaged in a promotional campaign for
18
In the story, two swindlers approached the Emperor,
falsely claiming the ability to make beautiful clothes from cloth
that could be seen only by those individuals fit for their positions
or who were not imbecils. The Emperor immediately hired them.
Word spread throughout the city about the unique quality of the
cloth and the personal characteristics that an individual must
possess to see clothes made of such material. After the swindlers
finished weaving the Emperor’s new clothes and presented them to
him, neither the Emperor nor his most trusted servants would admit
that they could not see the clothes for fear of appearing unfit or
stupid. Instead, each exclaimed that the clothes were beautiful.
Donning his new clothes, the Emperor walked in a procession
through the city’s streets. The townspeople also feared looking
stupid in their neighbors’ eyes. Like the Emperor and his servants,
they proclaimed that the clothes were the most beautiful they had
ever seen. It wasn’t until a child exclaimed, “But, Daddy, he has
nothing on!” that the crowd realized that the child spoke the truth.
39
Vioxx that minimizes the potentially serious
cardiovascular findings that were observed in the
[VIGOR] study, and thus, misrepresents the safety
profile for Vioxx. Specifically, your promotional
campaign discounts the fact that in the VIGOR
study, patients on Vioxx were observed to have a
four to five fold increase in myocardial infarctions
(MIs) compared to patients on the comparator
[NSAID], Naprosyn (naproxen).
Although the exact reason for the increased rate
of MIs observed in the Vioxx treatment group is
unknown, your promotional campaign selectively
presents the following hypothetical explanation
for the observed increase in MIs. You assert that
Vioxx does not increase the risk of MIs and that
the VIGOR finding is consistent with naproxen’s
ability to block platelet aggregation like aspirin.
That is a possible explanation, but you fail to
disclose that your explanation is hypothetical, has
not been demonstrated by substantial evidence,
and that there is another reasonable explanation,
that Vioxx may have pro-thrombotic properties.
...
Your minimizing these potential risks and
misrepresenting the safety profile for Vioxx raise
significant health and safety concerns. Your
misrepresentation of the safety profile for Vioxx
is particularly troublesome because we have
previously, in an untitled letter, objected to
40
promotional materials for Vioxx that also
misrepresented Vioxx’s safety profile.
...
We have idenitified a Merck press release entitled,
“Merck Confirms Favorable Cardiovascular Safety
Profile of VIOXX,” dated May 22, 2001, that is also
false or misleading for similar reasons stated above.
Additionally, your claim in the press release that VIOXX
has a “favorable cardiovascular safety profile,” is
simply incomprehensible, given the rate of MI and
serious cardiovascular events compared to naproxen.
The implication that Vioxx’s cardiovascular profile is
superior to other NSAIDs is misleading; in fact, serious
cardiovascular events were twice as frequent in the
VIOXX treatment group (101 events, 2.5%) as in the
naproxen treatment group (46 events, 1.1%) in the
VIGOR Study.
App. at 713-14, 718 (emphasis added).19
19
Also in the warning letter, the FDA identified specific
statements made by Merck in promotional audio conferences and
by Merck’s sales force demonstrating Merck’s minimization and
misrepresentation of the increased heart attack rates of Vioxx-
taking participants in the VIGOR study and several unsubstantiated
superiority claims made by Merck about Vioxx. App. at 715-16,
718-19. Finally, the warning letter concluded with a corrective
action plan which required Merck to issue a “‘Dear Healthcare
provider’ letter to correct false or misleading impressions and
41
The warning letter clearly and explicitly reprimanded
Merck for its (1) deceptive and misleading conduct in publicly
endorsing the naproxen hypothesis as the sole explanation for
the higher rate of cardiovascular events in VIGOR study
participants taking Vioxx, despite knowing that any purported
cardiovascular protective effect of naproxen was unproven, and
(2) downplaying of potential safety problems in failing to
disclose the possibility that Vioxx increases the risk of heart
attack. As the letter explained, this was not the first time the
FDA had charged Merck with misrepresenting Vioxx’s safety
profile. The language used in the letter was particularly strong
and indicated the FDA’s significant concern for the public’s
health. Also, the warning letter cannot be said to have
constituted mere speculation, but was rather a formal report of
“objective wrongdoing.” See Benak, 435 F.3d at 402(explaining that, in determining whether a plaintiff has inquiry
notice, “[s]peculation should not be given the same weight as
reports of objective wrongdoing”). Furthermore, the warning
letter was published on the FDA’s website where it would have
been discovered by a reasonable Merck investor. See In re
NAHC, 306 F.3d at 1325.
Moreover, the charges in the warning letter relate directly
to the misrepresentations and omissions alleged in the appellants’
complaint: that the company and certain of its officers and
directors intentionally misrepresented the cardiovascular safety
of Vioxx and, consequently, the impact that Vioxx would have
on Merck’s financial health. See DeBenedictis, 492 F.3d at 217-
information.” App. at 719.
42
18; see e.g., Amended Complaint, App. at 468 (stating that
“Defendants made... materially false and misleading statements
and omissions concerning... the safety profile of... VIOXX”);
App. at 470 (stating that “Defendants misrepresented the safety
profile of VIOXX, including concealing and minimizing the
significantly increased risk of heart attacks in patients taking the
drug”); App. at 482 (describing a “wrongful scheme... which
included the dissemination of materially false and misleading
statements and concealment of material adverse facts”); App. at
497 (stating that “Defendants falsely conditioned the market to
believe VIOXX was safe”). Accordingly, I believe that the
FDA’s warning letter to Merck sufficiently alerted a reasonable
investor to the possibility that Merck fraudulently misrepresented
the cardiovascular safety of Vioxx – its “blockbuster” product.20
20
It is important to note that Merck’s reliance on its
naproxen hypothesis was proved to be unfounded from the
beginning. Even before the FDA’s warning letter was issued, an
April 27, 2000, Reuters article reported that (1) a spokesperson for
leading naproxen manufacturer, Roche Holding Ltd., explained that
“[t]o [their] knowledge, naproxen does not prevent heart attack or
stroke” and (2) an ABN Amro analyst indicated that “[m]edical
authorities [he had] spoken to don’t see any special reduction of
such cardiovascular events in people taking naproxen.” App. at
2288. Additionally, an August 21, 2001, Bloomberg News Article,
reported a Merck representative’s comment that “[Merck] already
ha[s] additional data beyond what [the JAMA article] cite[s], and
the findings are very, very reassuring. VIOXX does not result in
any increase in cardiovascular events compared to placebo.” App.
at 539. Even if this “additional data” included evidence that could
43
Even assuming that the FDA’s warning letter alone did
not sufficiently excite “storm warnings,” the total mix of
information in the public realm which followed the warning
provided more than adequate “storm warnings” to put appellants
on inquiry notice.
In response to the FDA’s warning letter, there was
widespread media and financial analyst coverage commenting on
the FDA’s charges against Merck, with some reports noting that
such warnings are reserved for the more serious offenders. See
e.g., App. at 2353 (Reuters, September 24, 2001) (reporting that
“U.S. Regulators have charged... Merck... with misleading
doctors about its blockbuster painkiller Vioxx with promotions
that downplayed a possible risk of heart attacks”); App. at 2752
(Merrill Lynch, September 24, 2001) (stating that “[t]he FDA
issued a warning letter to Merck... [and] is looking for Merck to
cease all violative promotional activities... . We do not see how
this... can be helpful to Merck in promoting Vioxx”); App. at
2355 (USA Today, September 25, 2001) (reporting that “Merck’s
marketing efforts... have minimized Vioxx’s known and potential
cardiovascular risks, the FDA wrote in an eight-page ‘warning
letter’... . So far this year, the FDA has sent drug companies
fewer than a dozen warning letters, which the agency reserves for
activities that raise significant public health concerns”); App. at
2768 (UBS Warburg, September 25, 2001) (stating that the “FDA
[has] issue[d] [a] warning to Merck for marketing only one side
of the Vioxx safety argument... . Merck was cited several times
support Merck’s naproxen hypothesis, Merck never revealed the
details of its purported “additional data.”
44
for promoting the story that the outcome of the VIGOR study
was due to Naproxen being cardioprotective and that there is no
unusual cardiovascular safety risk with Vioxx.”); App. at 2360
(Associated Press, September 25, 2001) (reporting that “Merck
has argued that [the VIGOR study results make] Vioxx falsely
look[] risky because naproxen thins the blood... and thus
protect[s] against heart attacks... . ‘In fact, the situation is not all
that clear,’ [according to] the FDA”); App. at 2757 (Credit Suisse
First Boston, September 25, 2001) (stating that “the FDA [has]
issued a warning letter citing Merck with making misleading
statements in the promotion of... Vioxx”); App. at 2361 (The
Wall Street Journal, September 25, 2001) (reporting that
“Federal regulators warned Merck & Co. for improper marketing
of its blockbuster arthritis drug Vioxx, saying the company had
misrepresented the drug’s safety profile and minimized its
potential risks[,]” and “[w]hile the FDA sends out dozens of
routine citations annually, it issues only a handful of these more-
serious warning letters each year”); App. at 2363 (The New York
Times, September 26, 2001) (stating that “[t]he [FDA] has
ordered Merck & Company to cease promotions intended to
persuade doctors to prescribe its arthritis painkiller Vioxx, saying
the promotions minimize potential risks”). Even appellants
themselves recognized in their complaint that “FDA Warning
Letters are sent only to address serious circumstances.” App. at
1280.
Furthermore, in addition to the first lawsuit filed before
the FDA’s warning letter, three product liability and consumer
fraud actions had been filed in September and October 2001, all
alleging that Merck had misrepresented the cardiovascular safety
of Vioxx. See App. at 1748 (May 29, 2001, product liability
45
class action alleging that “Merck’s own research [demonstrated
that] users of Vioxx were four times as likely to suffer heart
attacks as compared to other less expensive medications..., [but
that] Merck... [took] no affirmative steps to communicate this
critical information to class members”); App. at 1557 (September
27, 2001, consumer fraud class action alleging that “Merck [had]
omitted, suppressed, or concealed material facts concerning the
dangers and risks associated with the use of Vioxx, including...
cardiovascular problems... [and] purposely downplayed and/or
understated the serious nature of the risks associated with
Vioxx”); App. at 1574 (September 28, 2001, product liability and
consumer fraud action alleging that Merck had “misrepresented
that Vioxx was... safe and effective..., when in fact the drug
causes serious medical problems such as an increased risk of
cardiovascular events, including strokes, heart attacks and
death”); App. at 1611 (October 1, 2001, product liability action
alleging that Merck failed to “[]disclose[]” that “Vioxx causes
heart attacks”). While these law suits did not allege securities
fraud, the general allegations contained within these complaints
relating to Merck’s intentional misrepresentation with regard to
Vioxx’s safety similarly formed the basis of appellants’
complaint.
Moreover, The New York Times article, dated October 9,
2001, quoted defendant Scolnick as explicitly stating that
“[n]aproxen lowers the heart attack rate, or Vioxx raises it.”
App. at 2367 (emphasis added). Based on my review of the
record, this express acknowledgment by a Merck representative
of the possibility that Vioxx actually raises the risk of heart attack
appears to be not only the first time such statement had been
made by the company, but also in stark contrast to Merck’s prior
46
representations. Therefore, because of what I perceive to be
significant media and financial analyst attention directed at the
explicit and serious nature of the FDA’s warning letter, the
allegations in the multiple lawsuits which followed, and Merck’s
change of tone in the October 9, 2001, article, I cannot see how
a reasonable investor could not be aware of the possibility that
Merck had been fraudulently misrepresenting the cardiovascular
safety of Vioxx.
Because the objective evidence indicated the possibility of
culpable activity on the part of Merck, a lack of significant stock
movement and decreases in analysts’ stock ratings and
projections do not negate a finding of “storm warnings” under
our inquiry notice standard. Appellants argue that “storm
warnings” could not have existed prior to the 2003 Harvard
Study because the total mix of public information did not have a
negative effect on the price of Merck stock or cause analysts to
drop their ratings for Merck or lower their projections for Vioxx
sales. It is true, as the majority points out, that our past inquiry
notice decisions have taken into consideration the market’s
response to disclosures alleged to constitute “storm warnings.”
However, I do not believe the law requires that, in order to make
a determination that “storm warnings” in fact exist, the total mix
of public information (purported to constitute “storm warnings”)
must have a negative effect on stock prices or cause analysts to
drop their ratings or lower their projections. See Benack, 435
F.3d at 400 (“information [need only suggest] possible
wrongdoing... to excite ‘storm warnings’”) (quoting In re NAHC,
306 F.3d at 1325) (emphasis added). As we recognized in In re
NAHC:
47
[S]torm warnings may take numerous forms, and
we will not attempt to provide an exhaustive list.
They may include, however, substantial conflicts
between oral representations of the brokers and the
text of the prospectus, ... the accumulation of
information over a period of time that conflicts
with representations that were made when the
securities were originally purchased, or any
financial, legal or other data that would alert a
reasonable person to the probability that
misleading statements or significant omissions had
been made.
306 F.3d at 1326 n.5 (quoting Mathews, 260 F.3d at 252 (internal
citations and quotations omitted)) (emphasis added). In my
view, fluctuations in stock price and analysts’ ratings and
projections, although relevant, are not a required consideration
in this circuit’s objective “storm warnings” analysis. Here, the
lack of a significant response from the market to the FDA’s
warning letter does not mean that the Emperor was not walking
down the street with no clothes on. It merely means that the
analysts saw the emperor’s new clothes as Merck described them
– not as reality presented.21
21
Regardless, Merck’s stock price did decline sharply in the
months leading up to October 9, 2001, as the public controversy
about Vioxx raged. From January 1, 2001, to October 9, 2001,
Merck’s stock price declined by $24.32 or 27.4% App. at 1770-73.
As appellants themselves alleged, “Merck’s stock price began its
slide in approximately January of 2001, and continued and
48
Based on the foregoing, I submit there were sufficient
“storm warnings” more than two years prior to the filing of
appellants’ complaint. At a minimum, I believe the FDA’s
September 17, 2001, warning letter constituted more than
sufficient “storm warnings” to put appellants on inquiry notice of
their claims, particularly since appellants fail to demonstrate
either that they conducted a diligent investigation within two
years of the accrual of such “storm warnings” or that they were
unable to uncover pertinent information during that time period.
Accordingly, because appellants waited over two years to bring
suit, I conclude that their claims were filed out of time and were
properly dismissed by the District Court.
worsened after August of 2001 when the VIGOR cardiovascular
data was presented more fully in the [JAMA article].” App. at
1225 (emphasis added).
49