LOCAL NO. 8 IBEW RETIREMENT PLAN & TRUST, on behalf of itself and all others similarly situated, Plaintiff, Appellant, v. VERTEX PHARMACEUTICALS, INC.; Joshua Boger, Ph.D.; Jeffrey Leiden, Ph.D.; Peter Mueller, Ph.D.; Paul Silva; Elaine Ullian; Nancy J. Wysenski, Defendants, Appellees, 838 F.3d 76

KAYATTA, Circuit Judge. During the course of clinical trials for an experimental drug combination intended to treat a fatal lung disease, Vertex Pharmaceuticals, Inc. (“Vertex”) announced interim results that overstated the improvement in lung function exhibited in a group of patients receiving the combination treatment. Following this announcement, Vertex’s stock price rose from $37.41 per share to close at $64.85 three weeks later. It then lost some of its gain, dropping to $57.80, after Vertex corrected the initial release’s overstatement.

Acting on behalf of all those who acquired Vertex stock during the period in which the overstatement stood uncorrected, Local No. 8 IBEW Retirement Plan & Trust (“Local No. 8”) filed this securities fraud class action complaint against Vertex and six past and current Vertex employees. The district court dismissed the complaint, finding that it failed to create a strong inference that the defendants had acted with scienter, the requisite mental state. See Local No. 8 IBEW Ret. Plan v. Vertex Pharm., Inc., 140 F.Supp.3d 120, 137 (D. Mass. 2015). We agree and so affirm. I.

Background As one of the world’s largest biotechnology companies, Vertex researches, develops, and sells treatments for a variety of ailments. In 1908, Vertex began working on drugs to combat cystic fibrosis, a fatal and as yet incurable lung disease. In early 2012 it gained Food and Drug Administration (“FDA”) approval to market a drug, Kalydeco, to treat patients with a rare form of the disease.

This approval, along with a contemporaneous drop in the value of Vertex’s stock due to Vertex’s diminishing returns from another product line, prompted Vertex to focus .its energies on developing a more broadly marketable cystic fibrosis treatment. In pursuit of this aim, Vertex explored a “combination therapy,” in which a cystic fibrosis patient first undergoes, a course of treatment with an experimental drug called VX-809 and only then begins taking Kalydeco.

Hoping that this combination wouíd be effective against the most common form of cystic fibrosis, Vertex began a three-phase clinical investigation required for FDA approval. See N.J. Carpenters Pension & Annuity Funds v. Biogen IDEC Inc., 537 F.3d 35, 39 (1st Cir. 2008) (describing the FDA approval process); 21 C.F.R. § 312.21 (describing the three phases of clinical investigation). On May 7, 2012, while the second phase of this process was ongoing, Vertex issued a press release announcing interim results drawn from roughly half of the 108 enrolled patients.

The press release focused in particular on one of the principal markers used to evaluate the effectiveness of a cystic fibrosis treatment: lung function, as measured by the amount of air a patient is capable of exhaling in one second. According to the press release, [o]f those who received [the combination therapy], approximately 46 percent (17/37) experienced an absolute improvement from baseline to Day 56 [of the tidal period] in lung function of 5 percentage points or more, and approximately 30 percent (11/37) experienced an absolute improvement from baseline to Day 56 of 10 percentage points or more.

None of the patients treated with placebo (0/11) achieved a 5-percentage point or more improvement from baseline to Day 56 in lung function. The press release described these results as “exceeding] [Vertex’s] expectations,” although it cautioned that “complete data” were not yet available and that “the final outcomes of this clinical trial or future clinical trials ... may be less favorable than the interim analysis reported today, or may not be favorable at all.” The same day, Vertex held a conference call for media and investors.

On the call, Vertex’s Executive Vice President and Chief Scientific Officer, Peter Mueller (“Mueller”), described the interim results as “really, really fantastic” and went on to say, “I have never seen anything like this.” Vertex’s Chief Executive Officer (“CEO”), Jeffrey Leiden (“Leiden”), also expressed confidence in the results, saying that they were “driving us to ... plan for potential market entries sooner than we had previously planned” and that, “[pjending final data this summer and discussions with regulators, we look forward to accelerating the development of our [cystic fibrosis] combination regimen.” Nancy Wysenski (“Wysenski”), at that time Vertex’s Chief Commercial Officer and Executive Vice President, further noted that the number of patients who stood to benefit from the combination treatment under review exceeded 70,000—a market that could translate into billions of dollars in potential sales.

Vertex’s stock price swiftly responded to the announcement of the promising interim results. On May 7, 2012, the day of the announcement, Vertex stock closed at $58.12 per share—up from the prior dose of $37.41, with a trading volume forty times higher than average. By May 25, 2012, the closing price had risen to $64.85 per share.

Meanwhile, five Vertex employees named as defendants in this suit— Joshua Boger (“Boger”), then Vertex’s Director; Paul Silva (“Silva”), who had formerly served as Vertex’s Vice President and Corporate Controller; Elaine Ullian (“Ullian”), Vertex’s co-lead independent director; Mueller; and Wysenski—sold a total of 539,313 shares of Vertex stock, collecting almost $32 million in all.

On May 29, 2012, Vertex announced in a press release that the interim results that had so energized its market prospects had overstated the improvement in lung function exhibited among the Phase 2 patients receiving the combination treatment. The error, as Vertex acknowledged that day in a conference call, stemmed from a “misinterpretation” as to whether the results Vertex had received from the third-party statistical analysis vendor reflected the absolute improvement in the patients’ lung function or, rather, the improvement relative to the patients’ baseline levels of lung function.

When evaluated properly, Vertex’s press release explained, the data showed that 35 percent of the patients taking the combination treatment (rather than 46 percent, as had initially been reported experienced an. absolute improvement of 5 percent or more, and that 19 percent (rather than 30 percent, as had initially been reported) experienced an absolute improvement of 10 percent or more.

Immediately following the announcement of the corrected results, the closing price of Vertex stock experienced its greatest decline in three years, dropping to $57.80 per share, down from $64.85 per share on May 25, yet. still well up from the May 4 close of $37.41. ■ Just short of two years later, Local No. 8 filed a class-action complaint against Vertex—as well as Boger, Leiden, Mueller, Silva, Ullian, and Wysenski—on behalf of all those who, like Local No. 8, had acquired Vertex stock between the announcement of the overstated interim results on May 7, 2012, and the announcement of the corrected results on May 29, 2012.

The complaint charged all defendants with securities fraud under

section 10(

b) of the Securities Exchange Act of 1934 (“Exchange Act”), 15 U.S.C. § 78j(b), and the Securities and Exchange Commission’s Rule 10b-5, 17 C.F.R. § 240.10b-5. It also charged the six individual defendants with joint and several liability under

section 20(

a) of the Exchange Act, 15 U.S.C. § 78t(a), for the alleged securities fraud, on the theory that these defendants “controlled Vertex, and/or controlled other Individual Defendants”; and charged Bol ger, .Mueller, Silva, Ullian, and Wysenski with insider trading under

section 20A of the Exchange Act, id. § 78t-l(a). The gravamen of the alleged fraud, according to the complaint, is'that, “[w]hen faced with ... study results' that seemed too good to be true, Defendants, rather than checking the results, turned a blind eye, accepting and promoting unlikely data that offered them a windfall on the sale of their stock.” The defendants moved to dismiss for failure to state a claim, see Fed. R. Civ. P, 12(b)(6), arguing that the facts alleged in the complaint fail to generate a strong inference that the defendants acted with the mental state required to render them liable under

section 10(

b) and Rule 10b-5. The district court agreed. It found as well that Local No. 8’s

section 20(

a) and

section 20A claims could not survive in the absence of a proper

section 10(

b) and Rule 10b-5 claim, and dismissed the complaint. See Local No. 8, 140 F.Supp.3d at 137. This timely appeal ensued. II. Analysis We review de novo the district court’s grant of the defendants’ motion to dismiss for failure to state a claim. Aldridge v. A.T, Cross Corp., 284 F.3d 72, 78 (1st Cir. 2002). A.

Section 10(

b) and Rule 10b-5 To successfully state a securities fraud claim under

section 10(

b) and Rule 10b-5, a plaintiff must adequately allege, among other things, scienter. “Scienter ... is 'a mental state embracing intent to deceive, manipulate, or defraud.’ ” Id. at 82 (quoting Ernst & Ernst v. Hochfelder, 425 U.S. 185, 193 n.12, 96 S.Ct. 1375, 47 L.Ed.2d 668 (1976)). A plaintiff can establish scienter “by showing that defendants either ‘consciously intended to defraud, or ... acted with a high degree of recklessness.’ ” Miss. Pub. Emps.’ Ret. Sys. v. Bos. Sci.

Corp. (“Boston I”), 523 F.3d 75, 85 (1st Cir. 2008) (quoting Aldridge, 284 F.3d at 82). “Recklessness in this context is ‘a highly unreasonable omission, involving not merely simple, or even inexcusable negligence, but an extreme departure from the standards of ordinary care.’” In re Smith & Wesson Holding Corp. Sec. Litig., 669 F.3d 68, 77 (1st Cir. 2012) (quoting Miss. Pub. Emps.’ Ret. Sys. v. Bos. Sci, Corp. (“Boston II”), 649 F.3d 5, 20 (1st Cir. 2011)).

The omission must “present[ ] a danger of misleading buyers or sellers that is either known to the defendant or is so obvious the actor must have been aware of it.” Id. (quoting Boston II, 649 F.3d at 20). This form of recklessness is “closer to a lesser form of intent” than it is to ordinary negligence, Greebel v. FTP Software, Inc., 194 F.3d 185, 199 (1st Cir. 1999).

To determine whether the complaint- here adequately alleges that the defendants acted with this culpable mental state, we eschew the ordinary standards of Federal Rule of Civil Procedure 8(a)(2), which require only that the plaintiff “plead[] factual content that allows the court to draw the reasonable inference that the defendants] [are] liable for the misconduct alleged,” Ashcroft v. Iqbal, 556 U.S. 662, 678, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009). Instead, Congress has directed us to evaluate

section 10(

b) and Rule 10b-5 claims of this type under the heightened pleading standards of the Private Securities Litigation Reform Act of 1995 (“PSLRA”), Pub. L. No. 104-67, 109 Stat. 737. See Aldridge, 284 F.3d at 78. As is relevant here, the PSLRA provides that a complaint must “state with particularity facts giving rise to a strong inference that the defendant[s] acted with [scienter].” 15 U.S.C. § 78u-4(b)(2)(A).

Under this standard, “[a] complaint will survive ... only if a reasonable person would deem the inference of scienter cogent and at least as compelling as any opposing inference one could draw from the facts alleged.” Tellabs, Inc. v. Makor Issues & Rights, Ltd., 551 U.S. 308, 324, 127 S.Ct. 2499, 168 L.Ed.2d 179 (2007). Local No/ 8 argues on-appeal that the complaint adequately alleges facts making it as likely as not that the defendants recklessly turned a blind eye to an obvious danger that, the announced

interpretation of the initial results was wrong. To support this argument, Local No. 8 points to the cumulative probative force of seven facts alleged in the complaint. We are mindful that “[e]ach individual fact about scienter may provide only, a brushstroke,” but it is our obligation to consider “the resulting portrait.” .In re Cabletron Sys., Inc., 311 F.3d 11, 40 (1st Cir. 2002).

Accordingly, we examine each alleged fact in turn, and then conclude by assessing them cumulatively. , First, Vertex itself described the results as unexpected, or as “exceeding] ... expectations.” The results were unexpected because, the complaint alleges, it was known “within Vertex” that VX-809 caused Kalydeco to “work less well.” The fact remains, though, that Vertex made the investment necessary to design and perform a' study testing the two drugs in combination. So, its puffing professions of surprise notwithstanding, Vertex must have' thought that positive results were possible, even if not probable.

We suspect, too, that many studies of new pharmaceutical products result in surprises, both good and bad. This moves us to Local No. 8’s second and related point, which arises from the complaint’s allegations concerning the science of cystic fibrosis research.

Local No. 8 alleges that cystic fibrosis research focuses on both “lung function and sweat chloride.” Because cystic fibrosis progressively and eventually fatally obstructs the lungs, “pulmonary function is an important marker of cystic fibrosis lung disease severity.” As a measure of pulmonary function, scientists test the patient for “Forced Expired Volume” (“FEV’). Local No. 8 alleges that scientists studying cystic fibrosis also measure “sweat chloride levels” because cystic fibrosis impairs the tissues of the sweat glands, thereby elevating the concentration of chloride in the patient’s sweat.

The interim results reported to and by Vertex showed increased FEV measurements, but no material drop in sweat chloride levels. Local No, 8 argues that some people have described sweat chloride levels to be the “gold standard” in cystic fibrosis research, and that “individuals at the Company were ‘highly skeptical’ of the study results because of the lack of sweat chloride improvements.” Therefore, reasons Local No. 8, it was obvious that there was something wrong with the results.

Missing from the allegations is any contention that any defendant viewed the sweat chloride levels as incompatible with the FEV measurements, or that any of the unnamed individuals conveyed any such skepticism to any defendant. Greebel, 194 F.3d at 199; cf. Auto. Indus. Pension Trust Fund v. Textron Inc., 682 F.3d 34, 39 (1st Cir. 2012) (inference that defendants suspected a statement was misleading is weaker where “warnings by subordinates or expressions of concern by executives are notably absent”).

The complaint does not even allege that scientists in general, much less those at Vertex, regarded the reported results as implausible. And given that the final results reflected the same phenomenon (improved FEV and steady sweat chloride levels), there is no reason given here to presume that scientists in general must view the possibility of such results as obviously wrong. Notably, too, Vertex reported the sweat chloride levels in the same press release in which it reported the positive FEV results.

So it would seem most likely that Vertex itself did not view the former as belying the latter, and neither apparently did the market. Third, Local No. 8 alleges that the study was very important to Vertex, and that it would therefore “be ‘absurd’ to suggest that Defendants were not aware of the suspect nature of the results.” It is true that the importance of a particular item to a defendant can support an inference that the defendant is “paying close attention” to that item. Institutional Inv’rs Grp. v. Avaya, Inc., 564 F.3d 242, 271 (3d Cir. 2009).

Such an inference, however, is only helpful in establishing scienter if that close attention would have revealed an incongruity so glaring as to make the need for further inquiry obvious. See id. at 270-71 (noting that a “steep decline” in operating margins creates inference that Chief Financial Officer, who “was paying close attention to these numbers,” would investigate the cause). We have already-discussed why the complaint fails to establish - that the announced results, on their face, contained such an obvious incongruity.

Similarly, some cases have recognizéd that certain key facts known to lower-level company managérs concerning a company’s flagship product, such as whether a $100 million contract has been signed to sell the product, or that sales are falling fast rather than rising, are very likely known to senior management who made repeated public announcements concerning sales of the product. See, e.g., Makor Issues & Rights, Ltd. v. Tellabs Inc., 513 F.3d 702, 706-10 (7th Cir. 2008).

But the complaint here does not allege that anybody at Vertex responsible for receiving, reviewing, and reporting the results had actually spotted the error in the

interpretation of the results before the discovery that led to the second announcement.

Fourth, Local No. 8 points to its allegation that the specific error in the publicly reported results—namely, the substitution of relative' improvement for absolute improvement in lung function— was so “fundamental” that it should have been apparent to the Vertex pul-monologist responsible for receiving the raw data from the' third-party vendor “regardless of how [the data] w[ere] presented by the vendor.” The pulmonol-ogist is not a defendant, however, and there is no allegation that any party responsible for the decision to announce the interim results received the raw data.

The fact that a Vertex pulmonologist was the one who received the raw data actually cuts sharply against Local No. 8 because there is no allegation that even this pulmonologist noticed or suspected that Vertex’s reported

interpretation of the results was incorrect, or told anyone of any skepticism. The complaint does assert in conclusory fashion that the pulmonologist “should have known” of the error. Yet, the complaint tells us nothing about the precise form of the. information conveyed by the vendor, or the vendor’s reliability. Negligence by the pulmonologist, too, hardly gets Local No. 8 anywhere. Rather, it adds a concrete .reason why the erroneous

interpretation of the, study results would not have been obvious to the executives to whom the pulmonologist reported the results.

Even making the reasonable inference, as Local No. 8 urges, that the defendants had access to the raw data—review of which would allegedly have rendered the error obvious through “simple math”—we have already determined that the complaint’s allegations are insufficient to establish that the erroneously interpreted end results (which are all the individual defendants are alleged to have received) were themselves so obviously suspect that we can draw a strong inference that the defendants were reckless in failing to consult the raw data themselves for verification.

Fifth, in its appellate brief, Local No. 8 also observes that it is “rare[ ]” to publish interim results and implies that Vertex’s decision to do so here is probative of scien-ter. However, the complaint nowhere alleges that the publication of interim results was anomalous, and so we do not consider this argument in assessing whether the complaint has stated a claim. Nor does Local No. 8 point to any legal requirement, or any undertaking by Vertex, that obligated the company to double-check the interim results before announcing them.

Sixth, we have considered Local No. 8’s allegations that the defendants had a financial motive to “turn[] a blind eye” to the erroneous

interpretation of the interim results because of the stock price spike precipitated by the error. Cf. Aldridge, 284 F.3d at 83 (“When financial incentives to exaggerate [material information] go far beyond the usual ..., they may be considered among other facts to show scienter.” (emphasis supplied)). Here, several facts strongly suggest that at least Vertex’s CEO, Leiden, had no motive to ignore an error that was obvious and that would therefore soon become known.

Leiden, who touted the erroneous interim results as driving Vertex “to acceleratfe] the development of [its] [cystic fibrosis] combination regimen,” is not alleged to have sold any stock during the class period. Local No. 8 contends that this was “a major study that ... was central to [Vertex’s] prospects.” Announcing good results on such a study would have been clearly better for Vertex than announcing great results only to reduce them to good results by shortly thereafter confessing error, thereby harming the company’s credibility and its reputation for competence.

Combined with the foregoing points, this fact makes it quite unlikely (and certainly less than 50-50) that any error was so obvious that Leiden must have known that the results were mistaken (and would therefore soon have to be withdrawn or corrected). Local No. 8 therefore places its focus on the fact that the other five defendants sold almost $32 million worth of stock following release of the overstated interim results. According to the complaint, these sales were “unusual when compared to [the individual defendants’] trading history before and after” the three-week class period.

Local No. 8 argues that this unusual activity, together-with the inferences that can be drawn from the defendants’ failure to double-check the interim results, makes “the inference that Defendants turned a blind eye to the suspect test results to line their own pockets ... at least as strong” as an inference of negligence. It is well settled that “[i]nsider trading in suspicious amounts or at suspicious times may be probative of- scienter.” Boston I, 523 F.3d at 92 (citing.Greebel, 194 F.3d at 197; Greenstone v. Cambex Corp., 975 F.2d 22, 26 (1st Cir. 1992)).

At the outset, however, it bears noting that, in addition to Leiden, defendant Boger did not engage in any inconsistent trading behavior during the class period. Boger, who was Vertex’s Director at the time, sold consistently small amounts of stock on a more or less weekly basis before, during, and after the class period.

So, to regard the stock sales as either motive for the fraud or evidence ,of the defendants’ knowledge that the interim study results had been misinterpreted, we must hypothesize either that the error was obvious. only to those defendants who made unprecedented sales, or that it was obvious to all, yet the Director and CEO nevertheless went along with announcing obviously flawed results. The complaint, though, offers no fact suggesting that the sellers knew more than the -nonsellers. To the contrary, the largest seller, Wysenski, was not a scientist.

And our discussion of Leiden’s salient interests and motive renders a stretch any inference that he would have gone along with announcing obviously erroneous results. The complaint’s chronology also offers a simple alternative explanation of the stock sales. After a long period of steady or dropping stock prices, the stock price suddenly jumped a large amount. Such an increase—no matter what its cause—creates a substantial incentive for holders to sell unless they believe the. price will continue to rise and are willing to wait.

Sales in the historical context described in the complaint carry little force in implying knowledge that the stock will drop. All in all, the presence of this perfectly understandable, innocent reason to sell, combined with the poor fit between the facts and Local No. 8⅛ theory, leaves us short of the scienter mark.

Seventh, there is, finally, the fact that Vertex announced the retirement of Wy-senski, aged fifty-four at the time, “suddenly and without any forewarning” on June 8, 2012—just one day after Iowa Senator Charles Grassley had sent a letter to Securities and Exchange Commission Chairwoman Mary Shapiro asking her to probe whether “Vertex ... executives ... took advantage of the spike in the stock knowing the news of the clinical data being overstated would be made public eventually, which in turn would negatively affect the stock value.” From these circumstances, Local No. 8 asks us to infer (1) that the defendants were aware of Senator Grassley’s letter, and (2) that the letter prompted Wysenski’s. retirement (3) because it correctly exposed that Wysenski, at a minimum, had deliberately turned a blind eye to the risk that the announced interim results were erroneous.

These allegations both point a finger at Wysenski and tend to exculpate the others who did not retire or leave the company. The question is whether they add enough to permit a claim against Wysenski; i.e., whether they make “the inference of scien-ter ... at least as compelling as any opposing inference one could draw from the facts alleged.” Tellabs, 551 U.S. at 324, 127 S.Ct. 2499. It is reasonable to infer that Vertex knew of Senator Grassley’s letter, and that Wysenski’s departure had something to do .with her stock sales. But -Local No. 8 must take us a step further.

For Local No. 8 to prevail, we would need to infer that Wysenski left or was pushed out for a particular reason; he., because she had unlawfully misled the market. If that were the reason, though, it would have applied to all defendants because the complaint makes no suggestion that Wysenski (a nonscientist handling marketing) knew anything more about the test results than did the others (who neither left nor were forced out in this timeframe). Alternative ■ explanations abound. Wy-senski’s very large sales and the spotlight focused on those sales could have given rise to her retirement without any hint of fraud.

Large insider sales by a senior manager, regardless of the reason for such sales, can present a major embarrassment for a company. Perhaps negligence by Wy-senski in preparing the erroneous press release prompted a forced retirement. Picking among these explanations, without the benefit of factual allegations suggesting Wysenski knew something the other defendants did not know, depends on a degree of guesswork inconsistent with the PSLRA pleading standard. Cumulatively, the brushstrokes here do not paint the required strong inference of scienter.

Vertex’s public description of a scientific study contained an error that made unexpectedly good results look' even better than they were; there is no claim that the pulmonologist who received and reviewéd the raw data behind the results noticed or reported the error to company executives; the company’s CEO, a scientist, had no plausible reason to announce results infected with an error that would most likely soon mar otherwise good news and harm Vertex by leading to an embarrassing correction; and there is no claim that the other defendants possessed any additional information.

Given the foregoing, the stock sales by some of the individual defendants and the timing of Wysen-ski’s retirement (which might otherwise look very different) cover too little canvas to evoke inferences of scienter strong enough to equal the alternative inference that Vertex was negligent in viewing very good results as being even better than they in fact were. Accordingly, the allegations underlying Local No. 8’s claim that the defendants acted with scienter fall short of what Congress demands in the securities fraud context. We therefore affirm dismissal of the

section 10(

b) and Rule 10b-5 claim. B.

Section 20(

a) and 20A Local No. 8 concedes that its remaining claims are derivative of its

section 10(

b) and Rule 10b-5 claim. Because we conclude that the district court properly dismissed the latter, it follows that the district court properly dismissed the former. We therefore affirm the dismissal of Local No. 8’s remaining claims. III. Conclusion Under the PSLRA, this action can only move forward if we find that the allegations make it at least as likely as not either that the defendants knew the results as reported were wrong, or that it was obvious to the defendants that they would discover the error if they looked.

Because we, like the district court, cannot so find, we affirm the dismissal of the complaint. . Because Local Ño. 8 appeals from a judgment granting the defendants’ motion to dismiss, we take the facts alleged in the complaint as true and draw all reasonable inferences from those facts in favor of Local No. 8. See In re Bos. Sci. Corp. Sec. Litig., 686 F.3d 21, 27 (1st Cir. 2012). . Phase 2 trials are typically closely controlled, small-scale studies designed to evaluate an experimental treatment's efficacy, as well as its short-term side effects and potential risks.

See 21 C.F.R. § 312.21(b). .Although the text of the press release was not incorporated into Local No. 8's complaint ■and was instead submitted by the defendants in support of their motion to dismiss, we may—as the district court did—nevertheless consider it at this stage because it is referenced in the complaint, it is central to Local No. 8’s claim, and no party disputes its authenticity. See Schaefer v. Indymac Mortg. Servs., 731 F.3d 98, 100 n.1 (1st Cir. 2013).

We also consider the subsequent press releases issued by Vertex on May 29, 2012 and June 28, 2012, both of which were submitted to the court below without objection. . Vertex's Chief Financial Officer confirmed in a call the following day that "the data [were] beyond our expectations” and that Vertex sought "to drive ... quickly into” the next phase of the clinical.investigation in order to “get to ... patients as fast as possible with this combination therapy.” .

According to the complaint, a relative improvement means "a percentage change from baseline,” whereas an absolute improvement is "the numerical distance between the baseline measurement and the improved measurement.” For our purposes, it appears that we need only understand the distinction to mqan that an absolute improvement is more favorable than a relative improvement of the same percentage. . Local No, 8 attempts to dilute this stringent standard by citing to a’ Ninth Circuit case, In re Oracle Corp. Sec.

Litig., 627 F.3d 376 (9th Cir. 2010), which stated that recklessness arises where the defendant “had reasonable grounds to believe material facts existed that were misstated or omitted, but nonetheless failed to obtain and disclose such facts al: though he could have done so without extraordinary effort,” id. at 390 (quoting Howard v. Everex Sys., Inc., 228 F.3d 1057, 1064 (9th Cir. 2000)). While the Ninth Circuit has indeed recently applied this formulation of recklessness in assessing the adequacy of a complaint under the PSLRA, Reese v.

Malone, 747 F,3d 557, 569 (9th Cir. 2014), even more recently it has rejected this same formulation as insufficient for assessing such a complaint, applying instead the formulation used in this circuit. In re NVIDIA Corp. Sec, Litig., 768 F.3d 1046, 1053 & n.7 (9th Cir. 2014), cert. denied sub nom. Cohen v. Nvidia Corp., — U.S. —, 135 S.Ct. 2349, 192 L.Ed.2d 143 (2015). In any event, however one might describe the law in the Ninth Circuit, we see no reason not to continue to apply a standard that makes clear that allegations- of merely unreasonable conduct do not sufficiently plead scienter. .

Local No. 8 contends that the district court improperly conflated the recklessness standard with an actual knowledge standard. We do not find the district court to have done so, but because we review the dismissal of Local No. 8’s complaint de .novo, under the proper standard, the outcome of this appeal in no way depends on our so finding, . Contrary to Local No. 8’s assertion, Boston I is not "particularly on point.” In Boston I, we reversed the dismissal of a

section 10(

b) claim, 523 F.3d at 94, finding that the defendants' own statements in connection with a manufacturing change to a medical device could be read as an admission that the change had been made in response to a design defect that the defendants had not previously disclosed, see id, at 88. No such potentially facially incriminating statements are at issue here. . At oral argument, counsel for Local No. 8 noted that, prior to discovery, few plaintiffs will be in a position to make specific allegations about the form of internal documents.

"But while a trawl through archives may sometimes catch, a few fish, Congress," for reasons of its own, "deliberately raised the entry bar to discovery ... through the PSLRA’s heightened pleading standards.” Textron, 682 F.3d at 40. . By contrast, consider a case of more or less steady stock price, followed by a nonpublic adverse event, inside sales, and then public disclosure of the adverse event. See, e.g., SEC v. Rocklage, 470 F.3d 1, 3-4 (1st Cir. 2006). . Local No. 8 cites a Ninth Circuit case, No. 84 Employer-Teamster Joint Council Pension Trust Fund v.

America West Holding Corp., 320 F.3d 920 (9th Cir. 2003), which it describes as "analogous.” However, a major factor in the America West court’s determination that the stock sales at issue were "unusual and suspicious,” id. at 940, was the fact that “[m]ost of the [alleged insiders] sold 100% of their shares, with the lowest percentage being 88%," id. at 939.

Here, by contrast, the two most senior defendants made no sales out of the ordinary course, and the complaint contains no allegation as to what proportion of his or her total stock any other defendant sold during the class period. .When asked about this point at oral argument, counsel for Local No. 8 stated, wrongly, that the complaint alleged that Senator Grass-ley's letter was publicly available, rather than asking us to take judicial notice of contemporaneous news reports establishing that the letter was indeed public.

See, e.g., Beth Healy, US Senator Charles Grassley Raises Vertex Stock Profits Issue with SEC Chief, Boston Globe, June 7, 2012, available at https://www, bostonglobe.com/business/2012/06/07/senator-charles-grassley-raises-vertex-stock-profits- ■ ‘ issue-with-sec-chief/cogtZEGDw2GhiDGt5 DNbIK/story.html, Because we do not find this point to be dispositive, we do not decide whether we would otherwise elect to take such notice sua sponte.