Tung v. Bristol-Myers Squibb Company

UNITED STATES DISTRICT COURT DOCUMENT

SOUTHERN DISTRICT OF NEW YORK ELECTRONICALLY FILED

DOC #:

JENNIFER TUNG, individually and on behalf of all others DATE FILED: 9/30/20 20

similarly situated,

Plaintiff,

1:18-cv-01611 (MKV)

-against-

OPINION AND ORDER

BRISTOL-MYERS SQUIBB COMPANY, MICHAEL GRANTING MOTION TO

GIORDANO, FOUAD NAMOUNI, FRANCIS M. CUSS, DISMISS

GIOVANNI CAFORIO, LAMBERTO ANDREOTTI, and

CHARLES A. BANCROFT,

Defendants,

MARY KAY VYSKOCIL, United States District Judge:

This is a securities class action seeking damages for alleged misstatements made by

Defendant Bristol-Myers Squibb Company (“BMS”) and its executives concerning a clinical trial

for a drug called Opdivo. In short, Plaintiffs accuse BMS of misleading the market about who

was eligible for the trial by defining eligibility criteria without regard for a purported industry

standard understanding of certain terms. It is alleged that as a result the market was led to

believe the trial was substantially more likely to succeed than in reality. After the trial failed,

BMS revealed to the market the full contours of the clinical trial, at which time some market

analysts and investors expressed surprise at BMS’s decisions regarding who was eligible for the

trial. This lawsuit soon followed. Following appointment of Lead Plaintiffs in this case, and an

initial round of briefing on a motion to dismiss, Lead Plaintiffs’ Consolidated Amended Class

Action Complaint (“CAC”) was dismissed for failure to plead scienter. Lead Plaintiffs were

granted leave to amend, and thereafter filed a Consolidated Second Amended Class Action

Complaint (“SAC”).1 Defendants again moved to dismiss. For the reasons that follow, and for

substantially the same reasons warranting dismissal of Lead Plaintiffs’ CAC, Defendants’ motion

is GRANTED.

BACKGROUND

The facts of this dispute are laid out in detail in the SAC, ECF #67, and in the opinion by

Judge Oetken granting Defendants’ motion to dismiss the previous complaint, ECF #66 (“CAC

Opinion”). The facts as stated herein are drawn from the SAC and are assumed true for the

purpose of this motion. See Littlejohn v. City of New York, 795 F.3d 297, 319 (2d Cir. 2015)

(“At the motion to dismiss stage, we accept these allegations as true and draw all inferences in

[Plaintiff’s] favor.”).

BMS is a pharmaceutical company. SAC ¶ 25. During the relevant class period—

January 27, 2015 to October 9, 2016—the company was led by executives including Defendants

Michael Giordano, Fouad Namouni, Francis M. Cuss, Giovanni Caforio, Lamberto Andreotti,

and Charles A. Bancroft. SAC ¶¶ 26-33.

Plaintiff’s allegations concern BMS’s development of the drug Opdivo, and specifically

the development of its use to treat non-small cell lung cancer (“NSCLC”). SAC ¶¶ 3-4, 7, 75.

Opdivo is a “checkpoint inhibitor.” SAC ¶¶ 39. Immune “checkpoints” are cellular mechanisms

that prevent the body’s immune system from attacking healthy cells. SAC ¶ 38. To treat

NSCLC, Opdivo targets the checkpoint protein PD-L1, which is expressed by cells in the body to

prevent the immune system from attacking them. SAC ¶ 38. Specifically, PD-L1 binds with the

“PD-1” protein on T-cells to prevent the T-cell from attacking the PD-L1 expressing cell. SAC

¶¶ 38. However, cancer cells can also express the PD-L1 protein, thereby preventing the body’s

1 By order dated May 16, 2018, the Arkansas Public Employees Retirement System and the Louisiana Sheriffs’

Pension and Relief Fund were appointed Lead Plaintiffs in this action.

immune system from attacking the cancer cells. SAC ¶ 39. By targeting cancer cells that are

expressing the PD-L1 protein, Opdivo allows the body’s immune system to attack the cancer

cells. SAC ¶ 38. Using the body’s own immune system to target and kill cancer cells in this

way is the goal of the burgeoning field of “immuno-oncology.” SAC ¶ 36.

The effectiveness of Opdivo and other checkpoint inhibitors depends on the level of PD-

L1 expression in a patient’s cancer cells (i.e. the share of cancer cells expressing the protein),

commonly expressed as a percentage. SAC ¶¶ 49-51. Generally, the greater a patient’s

expression of PD-L1, the more likely treatment with a checkpoint inhibitor will be effective.

SAC ¶ 3. It is estimated that approximately 70% of NSCLC patients exhibit some expression of

PD-L1 on cancer cells, while approximately 25% of NSCLC patients exhibit a 50% or higher

PD-L1 expression. SAC ¶ 3, 161. When developing trials to test the effectiveness of a PD-1

checkpoint inhibitor like Opdivo, companies like BMS have to determine a minimum expression

level a patient must exhibit to be included in the study. SAC ¶¶ 77, 79. In making this decision,

companies face a trade-off: the higher you set the cut-off, the more likely the trial will be

successful, but the lower the cut-off, the more patients who are eligible for the trial and,

ultimately, for any approved drug. SAC ¶¶ 10-11.

In January 2014, BMS announced a Phase III clinical trial for Opdivo called Checkmate-

026 (“CM-026”). SAC ¶ 77. The goal of CM-026 was to determine whether Opdivo

outperformed chemotherapy in treatment of NSCLC. SAC ¶ 77. If successful, CM-026 would

have been the last step before BMS could seek approval to market Opdivo for use to treat

NSCLC. SAC ¶¶ 75-77. BMS described that the purpose of CM-026 was “to show that

[Opdivo] will improve [outcomes] in subjects with strongly Stage IV or Recurrent PD-L1+ non-

small cell lung cancer when compared to chemotherapy.” (emphasis omitted). SAC ¶ 79. While

the trial was ongoing, in numerous statements, BMS also described the eligible class of

participants as exhibiting “strong” expressions of PD-L1. SAC ¶¶ 158-65. However, BMS also

repeatedly stated throughout the class period that it would not provide any further information

about its working definition of “strong.” See, e.g., SAC ¶¶ 93, 114.

Eventually, on August 5, 2016, BMS announced that CM-026 had failed to demonstrate

that Opdivo was more effective than chemotherapy. SAC ¶ 11. The same day, BMS

acknowledged for the first time that participants in CM-026 (i.e. those exhibiting “strong”

expressions of PD-L1) only needed to exhibit expressions of 5% or higher. SAC ¶ 164. The

price of BMS’s common stock fell approximately 16% from August 4 to August 5, 2016. SAC

¶ 209. About two months later, on October 9, 2016, BMS announced that the data from CM-026

was unable to provide any statistically significant conclusions whatsoever about the effectiveness

of Opdivo on NCSLC patients with expressions above 5%. SAC ¶ 13. On the next day of stock

trading, the price of BMS’s common stock fell approximately 10%. SAC ¶ 210.

This lawsuit was initiated in February 2018. See ECF #1. After Lead Plaintiffs were

appointed, the Parties briefed Defendants’ motion to dismiss the CAC. See ECF #51-53, 57, 62-

63. The CAC was dismissed by Judge Oetken, to whom this case was previously assigned,

based on Lead Plaintiffs’ failure to plead scienter. See CAC Opinion at 5-11. Lead Plaintiffs

were granted leave to amend and then filed the SAC. See ECF #67. The SAC seeks damages on

behalf of a class of investors in BMS’s common stock for BMS’s alleged misrepresentations

about the CM-026 trial targeting “strong” expressions of PD-L1. SAC ¶¶ 283-84. Lead

Plaintiffs suggest that applying the label “strong” to participants with only a 5% expression

contravened an industry standard understanding that an expression that low was understood to be

among the lowest “merely positive” expressions, and was certainly not a “strong” expression.

SAC ¶¶ 283-84. In support of this allegation, Plaintiffs point to previous studies, including a

roughly simultaneous clinical trial by BMS competitor Merck, which defined “strong”

expression as 50% expression. See SAC ¶¶ 85, 96.

Defendants again moved to dismiss the SAC. Defendants argue that Lead Plaintiffs do

not adequately allege scienter under the securities laws. See Memorandum in Support of Motion

to Dismiss SAC, ECF #72 (“Def. Br.”), at 8-20. This was the sole basis for Judge Oetken’s

previous dismissal of the CAC. The Defendants also argue here, however, that Lead Plaintiffs

have not identified a false or misleading statement of fact, see Def Br. at 20-24, and that loss

causation is not established. See Def. Br. at 24-25. Lead Plaintiffs opposed the current motion.

See Memorandum in Opposition to Motion to Dismiss the SAC, ECF #74 (“Pl. Opp.”).

Defendants then filed a Reply. See Reply Memorandum in Support of Motion to Dismiss, ECF

#76. The Court held oral argument on the motion on June 25, 2020. For the reasons that follow,

the Motion to Dismiss is granted.

LEGAL STANDARD

“To survive a motion to dismiss, a complaint must contain sufficient factual matter,

accepted as true, to ‘state a claim to relief that is plausible on its face.’” Ashcroft v. Iqbal, 556

U.S. 662, 678 (2009) (quoting Bell Atl. Corp. v. Twombly, 550 U.S. 544, 570 (2007)). “A claim

has facial plausibility when the plaintiff pleads factual content that allows the court to draw the

reasonable inference that the defendant is liable for the misconduct alleged.” Id. When

adjudicating a motion to dismiss, the Court “accept[s] all factual allegations in the complaint and

draw[s] all reasonable inferences in the plaintiff’s favor.” ATSI Commc’ns, Inc. v. Shaar Fund,

Ltd., 493 F.3d 87, 98 (2d Cir. 2007).

Here, Lead Plaintiffs assert claims under Sections 10(b), 20(a), and 20A of the Securities

Exchange Act of 1934 (“Exchange Act”). “Securities fraud claims are subject to heightened

pleading requirements that the plaintiff must meet to survive a motion to dismiss.” ATSI

Commc’ns, Inc., 493 F.3d at 99. The heightened pleading requirements are set forth in Rule 9(b)

of the Federal Rules of Civil Procedure and the Private Securities Litigation Reform Act of 1995,

Pub. L. No. 104-67, 109 Stat. 737 (“PSLRA”). Under the PSLRA, a plaintiff must “specify each

statement [or omission] alleged to have been misleading [and] the reason or reasons why the

statement is misleading.” See 15 U.S.C. § 78u-4. He or she must also “state with particularity

facts giving rise to a strong inference that the defendant acted with the required state of mind”—

i.e. the intent “to deceive, manipulate, or defraud”—with respect to each act or omission. Id.

“In a typical Section 10(b) private action a plaintiff must prove (1) a material

misrepresentation or omission by the defendant; (2) scienter; (3) a connection between the

misrepresentation or omission and the purchase or sale of a security; (4) reliance upon the

misrepresentation or omission; (5) economic loss; and (6) loss causation.” Stoneridge Inv.

Partners, LLC v. Scientific-Atlanta, Inc., 552 U.S. 148, 157 (2008). Relevant here, a plaintiff

can prove scienter by showing either “(1) a ‘motive and opportunity to commit the fraud’; or 2)

‘strong circumstantial evidence of conscious misbehavior or recklessness.’” Emps. Ret. Sys. v.

Blanford, 794 F.3d 297, 306 (2d Cir. 2015) (quoting ATSI Commc’ns, 493 F.3d at 99). “For an

inference of scienter to be strong [as required], ‘a reasonable person [must] deem [it] cogent and

at least as compelling as any opposing inference one could draw from the facts alleged.’” ATSI

Commc’ns, Inc., 493 F.3d at 99 (2d Cir. 2007) (quoting Tellabs, Inc. v. Makor Issues & Rights,

Ltd., 551 U.S. 308, 324 (2007) (alteration in original)).

ANALYSIS

A. Lead Plaintiffs Fail To Plead Scienter

In his opinion dismissing the CAC, Judge Oetken ruled that Lead Plaintiffs failed

adequately to allege scienter under either theory. See CAC Opinion at 6-11. First, he held that

the motives alleged by Lead Plaintiffs for the alleged fraud were not sufficient to prove scienter.

Id. at 6-7. In short, the Court ruled, a desire to “protect competitively sensitive information” was

ruled to be a non-actionable “generalized business motive.” Id. Judge Oetken also ruled that any

allegation that Defendants committed fraud “to sell stock while the price was artificially high”

failed because Lead Plaintiffs had not alleged sufficient facts to lead to that inference. Id. at 7.

Second, Judge Oetken also held that Lead Plaintiffs had not alleged “conscious misbehavior or

recklessness” because “[e]ven if the Court assume[d] the existence of a fixed usage of ‘strong’

that excluded a PD-L1 expression cut-off of 5%, the complaint still fail[ed] to plead facts giving

rise to a strong inference that the fixed usage ‘was either known to the defendants or so obvious

that the defendants must have been aware of it’ at the time of the alleged misrepresentations.”

See id. at 8 (citing ECA, 553 F.3d at 198).

In the SAC, Lead Plaintiffs abandon any argument that scienter is proven by Defendants’

motive to protect competitive information. Instead, Lead Plaintiffs focus their allegations on

attempting to prove the existence of an industry-wide consensus on the meaning of “strong” in

the context of PD-L1 expression. See Pl. Opp. at 9-15. They also argue briefly that motive can

be inferred from the insider trading allegations in the SAC regarding stock trades by BMS

executives. See Pl. Opp. at 15-17. Each of these arguments fails.

i. Conscious Misbehavior Or Reckless Conduct

Judge Oetken held, and, thus, it is law of the case that, Lead Plaintiffs cannot prove the

existence of an industry standard simply by pointing to the definition of “strong” used by Merck,

BMS’s competitor who was running a simultaneous PD-L1 checkpoint inhibitor study. See CAC

Opinion at 10; see also Johnson v. Holder, 564 F.3d 95, 99 (2d Cir. 2009) (“The law of the case

doctrine commands that ‘when a court has ruled on an issue, that decision should generally be

adhered to by that court in subsequent stages in the same case’ unless ‘cogent and compelling

reasons militate otherwise.’” (quoting United States v. Quintieri, 306 F.3d 1217, 1225 (2d Cir.

2002))). Judge Oetken found that, at best, such allegations could prove that “Merck’s definition

of ‘strong’” was greater than 5%, but that it did not determine any industrywide definition. Id.

He stated that Lead Plaintiffs would need to cite a “publication, guidance, communication,

document, or other specific source of information that could have alerted Defendants, at the time

of the alleged misrepresentations, to an industrywide consensus that ‘strong’ PD-L1 expression

was inconsistent with a 5% cut-off.” Id. Only if an industry standard were established, could

Plaintiff’s cogently allege Defendants’ conscious misbehavior or recklessness sufficient to

establish scienter.

In response to that instruction, Lead Plaintiffs added three types of allegations to the

SAC. First, Lead Plaintiffs have included statements and conclusions from a purported expert

witness, Ronald H. Blum, a medical oncologist and researcher who has “participated in the

independent oversight of late phase clinical trials, including clinical trials examining PD-L1

checkpoint inhibitors.” SAC ¶ 9. Dr. Blum states that by the start of the class period in this case

(January 27, 2015), there was an industry-wide consensus that 5% expression of PD-L1 was

“low or minimal” and that 50% expression was “strong.” SAC ¶ 9. Second, the SAC includes

statements from a series of confidential witnesses who were former employees at BMS. The

former employees explain that BMS executives were aware of Merck’s usage of a 50% threshold

as “strong” and that BMS was using a different number. SAC ¶¶ 71-72, 86-90. Finally, the SAC

also includes certain other circumstantial evidence of wrongdoing that Lead Plaintiffs claim lead

to an inference of scienter, such as BMS’s stated intent to “harmonize PD-L1 assays” and the

departure from BMS of certain individual Defendants. SAC ¶¶ 233, 279-80. Each of these new

pieces of information is insufficient to remedy the deficiency in the scienter charge identified in

Judge Oetken’s opinion.

It is settled in this Circuit that plaintiffs may rely on testimony of confidential witnesses

to bolster their complaint, provided that the witnesses are described “with sufficient particularity

to support the probability that a person in the position occupied by the source would possess the

information alleged.” Novak v. Kasaks, 216 F.3d 300, 309 (2d Cir. 2000). Lead Plaintiffs’

confidential sources are clearly described with sufficient detail in the SAC, which notes their

roles at BMS and to which of the Defendants each spoke. See SAC ¶¶ 71-72, 87-90. However,

Lead Plaintiffs’ problem is not with the reliability of the sources but with what information they

provide. The former employees claim only that BMS was aware of Merck’s operating

definitions and that BMS would have incorporated that information into “forecasting to

understand the market.” SAC ¶ 72. This information does not add anything new to the case.

Lead Plaintiffs previously alleged in the CAC that BMS was aware of Merck’s definition, and

Judge Oetken held that knowledge was insufficient to prove anything other than Merck’s

definition of “strong.” See CAC Opinion at 10.

The inference of scienter urged by Lead Plaintiffs is not “cogent” or “at least as

compelling” as the opposing inference urged by Defendants. Tellabs, Inc. v. Makor Issues &

Rights, Ltd., 551 U.S. 308, 324 (2007). Plaintiffs’ unsupported allegations of BMS’s knowledge

and disregard of an industry standard are less convincing than Defendants’ inference that even if

BMS executives were aware of Merck’s thresholds, they used a different definition in an attempt

to reach a broader segment of cancer patients. Importantly, none of Lead Plaintiffs’ sources can

verify that BMS believed that Merck’s thresholds represented an “industry standard” definition.

Contrary to Lead Plaintiffs’ contentions, see Pl. Opp. at 10-11, requiring proof from the

confidential sources of BMS’s direct knowledge of an alleged industry standard is not “smoking

gun” evidence beyond what they are required to provide. See Tellabs, 551 U.S. at 324. Rather,

that missing information is necessary to support Lead Plaintiffs’ suggested inference that

Defendants were anything more than negligent in their design and execution of the CM-026 trial.

Plaintiff’s reliance on Dr. Blum’s testimony fares no better. A plaintiff can rely on expert

testimony to bolster the factual allegations of its complaint, but the Court may not consider any

legal conclusions offered by the expert. See, e.g., Ong v. Chipotle Mexican Grill, Inc., 294 F.

Supp. 3d 199, 224 (S.D.N.Y. 2018) (striking legal conclusions about Defendant’s

misrepresentations but considering expert’s testimony about industry practices). Here, Dr.

Blum’s conclusions are proper except to the extent he opines that Defendants’ statements were,

in fact, misleading. See SAC ¶¶ 120-21, 137.

Dr. Blum concludes that an industry standard definition of “strong” and “low” PD-L1

expression existed by the beginning of the class period. See SAC ¶ 9. Leaving aside that his

conclusion may be refuted by other evidence in the SAC, see infra at 12, and even accepting as

true his assertion that such a standard existed in some or all of the immuno-oncology field,

Plaintiff has provided evidence of that standard only through Dr. Blum. Failure to provide some

other industry publication, guidance, or other evidence of such a standard, in addition to the

expert testimony, is legally insufficient and requires that the Court not rely on Dr. Blum’s

testimony at all. See Ong, 394 F. Supp. 3d at 223 (“‘[E]ven if non-opinion portions of an

expert’s affidavit constitute an instrument pursuant to Rule 10, opinions cannot substitute for

facts under the [PSLRA].’” (quoting Fin. Acquisition Partners LP v. Blackwell, 440 F.3d 278,

285-86 (5th Cir. 2006))); accord In re CommVault Sys. Inc. Sec. Litig., No. 14-cv-5628 (PGS),

2016 WL 5745100, at *4-7 (D.N.J. 2016) (considering expert testimony only when plaintiff

“d[id] not rely exclusively” on the expert) (cited at Pl. Opp. at 12).

The critical deficiency in Lead Plaintiffs’ case is that even if the Court were to consider

Dr. Blum’s testimony, he states nothing that allows an inference that BMS executives had

knowledge of a new industry standard definition. Dr. Blum does nothing to bring his claim of an

industry standard home to BMS. Instead, the SAC makes the conclusory allegation that Dr.

Blum’s understanding of an industry standard “should be imputed to Defendants.” SAC ¶ 16.

As Defendants’ explain, Lead Plaintiffs have not alleged how or when an industry understanding

reached BMS. See Def Br. at 13. This is particularly important since no one has suggested that

that Dr. Blum has ever worked at BMS or is familiar with its procedures and practices. Without

more, the Court cannot find that Dr. Blum’s testimony establishes the existence of an industry

standard or, more importantly, knowledge of such a standard at BMS. As a result, Defendants

cannot be said to have knowingly or recklessly made statements which contradicted the putative

standard.

The other information added to the SAC is circumstantial evidence that does nothing to

further an inference of scienter. Lead Plaintiffs’ note, for example, that BMS, in June 2015,

committed that they “will participate in the efforts to harmonize PD-L1 expression assays.”

SAC ¶ 233. This statement actually supports the inference that no industry standard definitions

existed in June 2015, as BMS was committing to “harmonize” approaches at that time, and not

recognizing that they had already done so. As another example, Lead Plaintiffs bolstered their

allegations regarding the “suspicious departures” of Defendants Cuss and Giordano from BMS.

See SAC ¶¶ 279-80. Judge Oetken previously dismissed these allegations as not sufficient to

plead scienter on their own. See CAC Opinion at 10 n.4. Because none of the new allegations

add anything to the scienter analysis, the allegations remain insufficient. They hardly give rise to

an inference of scienter that is “cogent and at least as compelling as any opposing inference”

offered by Defendants, Tellabs, Inc., 551 U.S. at 324, who persuasively argue that the departures

were in the normal course of business and that BMS otherwise was not contradicting as as-yet-

undeveloped industry-wide understanding of PD-L1 expressions.

Finally, the SAC incorporates by reference at least one document that seems to

contradict Lead Plaintiff’s arguments about the existence of an industry standard. See Roth v.

Jennings, 489 F.3d 499, 509 (2d Cir. 2007) (“In certain circumstances, the court may permissibly

consider documents other than the complaint in ruling on a motion under Rule 12(b)(6).

Documents that are attached to the complaint or incorporated in it by reference are deemed part

of the pleading and may be considered.”). Specifically, in support of the argument that other

industry participants adopted the 50% definition of “strong” expression of PD-L1, the SAC

references industry presentations by BMS competitors and other groups using that terminology.

See SAC ¶ 246.

In particular, the SAC mentions an October 8, 2015 presentation by Nektar, a

pharmaceutical research company, at an industry conference in which the company “spoke about

[Merck’s PD-L1 inhibitor’s] application for the ‘subset of high PD-L1 expression of more than

50%.’” SAC ¶ 246. Defendants submitted with their motion papers a slide from that

presentation, which occurred almost two years after CM-026 began and almost one year into the

class period. See Reply Declaration of Matthew Solum in Support of Motion to Dismiss, ECF

#76, Ex. 27 (the “Nektar Slide”) at 3. That slide includes four different definitions of PD-L1

“positive” and two different definitions of “strong.” See Nektar Slide at 3. Far from supporting

the inference that an industry standard existed before the class period, the slide supports

Defendants’ argument that almost two years into the CM-026 trial, the industry had not yet

settled on any standard, and thus that Defendants could not have knowingly misrepresented that

its trial conformed to such a standard. Considering all of the allegations together, Tellabs, Inc.,

551 U.S. at 323, the Court again concludes that Lead Plaintiffs have not alleged a compelling

inference of scienter based on Defendants’ conscious misbehavior or reckless conduct.

ii. Fraudulent Motive And Opportunity

While Lead Plaintiffs largely devote their scienter arguments to proving “conscious

misbehavior or recklessness,” they have included some additional allegations of insider trading

in the SAC, in an effort to prove Defendants had a fraudulent motive to mislead investors. While

Judge Oetken previously rejected a version of this argument, see CAC Opinion at 6-8, the SAC

adds new allegations about the trades. In particular, Lead Plaintiffs now allege that the

Individual Defendants, most notably Andreotti, sold shares of BMS common stock before the

failure of CM-026 was disclosed, resulting in a total profit of over $55 million. SAC ¶ 264.

Even with the additions, however, these allegations do not rise to a level sufficient to lead to a

strong inference of scienter.

Insider trading activity proves motive if the executive, who possesses information that the

market does not, makes “unusual” trades in a way to avoid the ultimate loss due to the public

disclosure of that information. See In re Scholastic Corp. Secs. Litig., 252 F.3d 63, 74 (2d Cir.

2001). Unusual trades occur when an executive profits highly from sales, sells a large portion of

his or her stock holdings, or trades drastically more stock during the class period than he or she

did before. Id. at 7475. Here, Defendants convincingly argue, see Def. Br. at 9-10, that BMS

executives sold the same overall percentage of the stock they owned before and during the class

period. See Def. Br. at 9-10. Defendants further clarify that while executives experienced

greater profits from stock sales, these were a result of Defendants receiving more shares as

compensation overall during the class period and not opportunistic trading, since the same

proportion was sold. See id. This uniform trading pattern, along with the fact that the vast

majority of the trades in question were made pursuant to Rule 10b5-1 stock plans, defeats any

inference that Defendants were motivated by profits. See In re Lululemon Sec. Litig., 14 F.

Supp. 3d 553, 585 (S.D.N.Y. 2014), aff'd, 604 F. App'x 62 (2d Cir. 2015) (“Trades made

pursuant to a Rule 10b5-l trading plan do not give rise to a strong inference of scienter.”).

Finally, Lead Plaintiffs have omitted information regarding the Defendants’ total holdings. That

omission is fatal to establishing scienter based upon trading activity. See In re eSpeed, Inc. Secs.

Litig., 457 F. Supp. 2d 266, 290 (S.D.N.Y. 2006).

Lead Plaintiffs are correct, however, that certain of the trades stand out. Namely, in May

2016, about two months after Defendants again claimed that CM-026 participants had a “strong”

PD-L1 expression, Defendants Anderotti, Bancroft, and Caforio all sold significantly greater

numbers of shares than previously had been their practice. See SAC ¶ 261. None of these

trades were made pursuant to a Rule 10b5-1 plan. SAC ¶ 261. The trades came at a pivotal time

during the CM-026 trial, after again describing the trial to the public, but shortly before the

results were revealed. At argument, counsel for Defendants attempted to distinguish these trades

on the basis that the results of CM-026 were not yet known. See Tr. of June 25, 2020 Oral

Argument at 14:23-16:16.

While these trades are notable, in light of Lead Plaintiffs’ pleading failures outlined

previously, they cannot by themselves give rise to a strong inference of scienter. See Tellabs,

Inc., 551 U.S. at 324. This is especially true since the SAC alleges only three of the Individual

Defendants made trades at that time, as Judge Oetken previously noted. See CAC Opinion at 7

(collecting cases); see also San Leandro Emergency Med. Grp. Profit Sharing Plan v. Philip

Morris Cos., Inc., 75 F.3d 801, 814 (2d Cir. 1996) (“[T]he fact that other defendants did not sell

their shares during the relevant class period sufficiently undermines plaintiffs’ claim regarding

motive.”). In short, Lead Plaintiffs have not provided sufficient facts to support a strong

inference of scienter based on motive and opportunity to commit the fraud. As a result, Lead

Plaintiffs’ claims cannot stand.

B. Lead Plaintiffs Also Fail To Plead A Material Misstatement Or Omission

While Judge Oetken’s CAC Opinion did not consider whether Lead Plaintiffs adequately

pleaded false or misleading statements or omissions, the Court also concludes that the statements

in the SAC are not material misrepresentations or omissions required for Section 10(b) liability.

See Stoneridge Inv. Partners, LLC, 552 U.S. at 157 (“In a typical § 10(b) private action a

plaintiff must prove (1) a material misrepresentation or omission by the defendant . . .”). As

Lead Plaintiffs describe it, the SAC alleges three types of misstatements or omissions: (1) claims

that CM-026 focused on “strong” PD-L1 expressions, when it did not, (2) failure to state exactly

what the cutoff for CM-026 was, despite BMS noting from the outset that it would not do so, and

(3) statements about BMS’s confidence in the success of CM-026. See generally SAC ¶¶ 166-

205. None of these types of statements are actionable.

First, BMS’s decision to describe CM-026 as “strong” was not misleading. As described

above, Lead Plaintiffs have not alleged facts sufficient to suggest that there was an industry

standard definition of that term. Without such a definition, or indication that Defendants

themselves agreed with Lead Plaintiffs’ supposed definition before speaking, the statements are

not actionable. See, e.g., In re Eros Int’l Secs. Litig., No. 15-cv-8956 (AJN), 2017 WL 6405846,

at *5 (S.D.N.Y. Sept. 22, 2017) (“In the absence of a shared definition of the term . . . Plaintiffs

must identify affirmative statements made by Defendant . . . that can plausibly be construed as

false or misleading.”). And, as Lead Plaintiffs themselves allege, throughout the class period,

BMS declined to offer publicly any definition of the term “strong” as used in reference to the

CM-026 clinical trial. SAC ¶¶ 93, 114. Moreover, the Nektar Slide discussed above indicates

that in October 2015, industry sources both were aware that BMS’s definition of “strong” was

greater than five percent and that no industry standard definition of the term otherwise existed.

This defeats any argument that Defendants’ use of “strong” to refer to PD-L1 expression in CM-

026 participants was false in light of a supposed industry standard.

Second, Defendants’ failure to disclose the actual cutoff for CM-026 also is not

actionable. Lead Plaintiffs argue that this omission is actionable only because it would lead

investors to believe that BMS was targeting a different category of patients than they actually

were. Pl. Opp. at 20. However, this argument also depends on the existence of a purported

industry standard understanding of “strong” when used in reference to the expression of PD-L1.

Simply, if no industry-wide understanding exists, there is no reasons for investors to conclude

anything in particular about CM-026. As discussed, Lead Plaintiffs have failed to allege such an

industry standard. Lead Plaintiffs also assert that “Defendants’ adoption of ‘the exact same

terms (e.g. ‘strong’ and ‘high’) used by Merck . . . was ‘highly unusual and misleading’ and

‘disingenuous.’” Pl. Opp. at 20 (quoting SAC ¶¶ 80-81, 97). However, in the absence of an

industry standard definition, no reasonable investor would have any particular understanding of

those terms in light of Merck’s usage, and Lead Plaintiffs certainly have not alleged

particularized reasons why those descriptions would be materially misleading.

Lead Plaintiffs also assert that, even if Defendants had no duty to disclose the true cutoff

for CM-026 at its onset, a duty to disclose developed during the trial. See Pl. Opp. at 23-24. In

support of this assertion, Lead Plaintiffs note that “once a company speaks on an issue or topic,

there is a duty to tell the whole truth.” Menaldi v. Och-Ziff Capital Mgmt. Grp. LLC, 164 F.

Supp. 3d 568, 583 (S.D.N.Y. 2016). Unlike the Menaldi case, however, Lead Plaintiffs have not

alleged that Defendants gave any incomplete statements. Indeed, BMS fully described CM-026

using the terms as BMS understood them. No industry standard existed before the class period

here, and Lead Plaintiffs do not attempt to argue that an industry standard emerged during the

class period.2

Finally, the other statements alleged in the SAC are non-actionable forward-looking

statements or opinion. The Second Circuit has been clear that courts should not engage in post

hoc analysis of a company’s optimistic statements, but instead should look to whether plaintiff

pleads facts sufficient to allege that the optimism was misleading. Kleinman v. Elan Corp., plc,

706 F.3d 145, 154–55 (2d Cir. 2013). Here, Lead Plaintiffs have failed to allege an industry

standard understanding of “strong” that would have been inherent in any BMS executive’s

representations about the likelihood of success of CM-026. Instead, Lead Plaintiffs take issue

only with statements that express an optimism about the future of BMS’s business. Those are

not actionable. In re Sanofi-Aventis Sec. Litig., 2009 WL 3094957, at *4 (S.D.N.Y. Sept. 25,

2009) (“It is well settled that most forward-looking statements, such as predictions about future

revenue, accompanied by cautionary language are not actionable under Rule 10b-5.”).

In sum, largely for the same reasons that Lead Plaintiffs have not alleged scienter—

because they have failed to establish an inference that there existed an industry standard

understanding of a “strong” PD-L1 expression—the statements Lead Plaintiffs identify in the

SAC are not actionable for Section 10(b) purposes.

C. Lead Plaintiffs’ Claims Pursuant to Sections 20(a) and 20A Also Must Be Dismissed

The SAC includes two causes of action against the Individual Defendants, alleging control-

person liability under Sections 20(a) and 20A of the Exchange Act. To establish a prima facie

case of control-person liability however, a plaintiff must successfully allege a primary violation.

2 The Court once again notes that the Nektar Slide, which was used at an industry conference in October 2015,

almost two years into CM-026 and almost one year into the class period, stated that Merck and BMS had different

definitions of “strong” and that the industry as a whole had no settled definition of PD-L1 positivity, with at least

three different definitions reflected on that slide. See Nektar Slide at 3. Lead Plaintiffs assert that an industry

standard was developed before the class period began, but the Nektar Slide alone demands a different conclusion.

See ATSI Commce’ns, Inc., 493 F.3d at 108. Because the Court concludes that Lead Plaintiffs’

have not alleged scienter or a false statement, essential elements of a 10(b) claim, they have

failed to allege any primary violation. As a result, the claims under Sections 20(a) and 20A also

fail to state a claim and, accordingly, are dismissed.

CONCLUSION

Following dismissal of the First Amended Consolidated Class Action Complaint, Lead

Plaintiffs’ attempted to remedy the specific deficiencies identified in the Dismissal Order. Those

efforts, as described herein, fall short. The Second Amended Complaint suffers from the same

deficiencies as the first. Namely, Lead Plaintiffs have not alleged sufficient particularized facts

to lead the Court to a strong inference of scienter as required by the PSLRA. The complaint fails

to establish the existence of an industry standard definition relating to PD-L1 expression, as

would be required to find that Defendants knowingly or recklessly disregarded the standard.

Moreover, Lead Plaintiffs’ allegations of insider trading, while in one small part notable, are

collectively not sufficient to establish an inference of motive to commit fraud.

Additionally, Lead Plaintiffs’ failure to allege the existence of an industry standard

definition means that they also have failed to allege any materially misleading statement or

omission.

As such, Defendants’ Motion to Dismiss the Second Amended Consolidated Complaint is

GRANTED and the complaint is DISMISSED WITH PREJUDICE. The Clerk of Court

respectfully is requested to close the case.

SO ORDERED.

Jw K V yehencl

Date: September 30, 2020 MARY KAY VYSHOCIL

New York, NY United States District Judge

18