Di Donato v. Insys Therapeutics Incorporated

• Court: District Court, D. Arizona
• Decided: September 20, 2019
• Docket: 2:16-cv-00302
• Status: Unknown

Opinion

1 WO 2 3 4 5 6 IN THE UNITED STATES DISTRICT COURT 7 FOR THE DISTRICT OF ARIZONA 8 Richard Di Donato, individually and on No. CV-16-00302-PHX-NVW 9 behalf of all others similarly situated, ORDER 10 Plaintiff, 11 v. 12 Insys Therapeutics, Inc.; Michael L. 13 Babich; Darryl S. Baker; John N. Kapoor; and Alec Burlakoff, 14 Defendants. 15 16 17 Before the Court are Lead Plaintiff’s Motion for Class Certification (Doc. 159) and 18 Defendants’ Motion for Leave to File Sur-Reply in Further Opposition to Lead Plaintiff’s 19 Motion for Class Certification (Doc. 170). 20 On June 10, 2019, Defendant Insys Therapeutics Inc. (“Insys”) commenced a 21 voluntary case under chapter 11 of title 11 of the United States Code (11 U.S.C. § 101 et 22 seq.) in the United States Bankruptcy Court for the District of Delaware, and this action is 23 automatically stayed with respect to Insys. (Doc. 230.) This action is not stayed as to the 24 proceedings against the Defendants other than Insys. (Doc. 258.) This Order, therefore, 25 applies only to Defendants Michael L. Babich, Darryl S. Baker, and John N. Kapoor. 26 Claims against Defendant Alec Burlakoff were dismissed. (Doc. 107 at 41.) 27 28 1 I. BACKGROUND 2 Insys is a publicly traded1 pharmaceutical company headquartered in Arizona that 3 makes and sells Subsys, a very expensive sublingual fentanyl spray approved by the Food 4 and Drug Administration to treat breakthrough pain in adult cancer patients already taking 5 opioid pain medication. Babich, Baker, and Kapoor served as high-level executives at 6 Insys during 2014 through 2016. During that time, Insys reported increasing revenues each 7 quarter and attributed most of its upward trajectory to the sale of Subsys. Insys touted its 8 unique programs within the oncology setting for increasing sales and claimed that it was 9 taking market share from competing products for breakthrough cancer pain by building 10 awareness among oncologists. However, as alleged, Insys actively discouraged promoting 11 Subsys for cancer patients, less than 1% of Subsys prescriptions were written by 12 oncologists, and Insys paid doctors bribes and kickbacks in exchange for prescribing 13 Subsys off-label. 14 Company executives publicly attributed the drug’s success to vigorous marketing 15 efforts and the company’s diligent negotiations with pharmacy benefit managers, third- 16 party payers, and insurers. Insys created an “Insurance Reimbursement Center,” 17 sometimes called the “Prior Authorization Unit,” ostensibly to interact with third-party 18 payers to obtain prior authorization and insurance coverage for Subsys prescriptions. 19 Employees of the “Insurance Reimbursement Center” allegedly falsified patient diagnoses, 20 lied to insurance companies, and posed as employees of the prescribing doctors. 21 Plaintiff alleges that Defendants made the following materially false and misleading 22 statements: 23 268. More specifically, Defendant Babich stated: 24 I think Q4 is a great indication of what we can do with the product moving forward, as well. I think that is important for 25 folks—our sales force expansion was based on opportunity. 26 We keep hitting new highs in the number of new doctors that we activate on a weekly basis. We have some very unique 27 28 1 During the Class Period, Insys common stock traded on the NASDAQ.

-2- programs within the oncology setting that we continue to 1 execute on and any growth that we see in this overall TIRF 2 class is specifically coming from Subsys. 3 So we feel that this is our market to continue to grow and to continue to dominate, like we are doing at this point with our 4 market share. I’ve always talked about, from a market share, 5 our next total is 50% market share. You can see that in Q4 the Actiq generic continued to decline, so we continued to take 6 market share from the generic. And I think that’s a testament 7 to the fact that we have a clinically superior product to the Actiq generic out there. So I think long term we can 8 eventually get to that 60% market share for this product. 9 .... 10 272. Furthermore, with respect to the continued growth of Subsys, the FY14 Form 10-K also explained: 11 some of the key factors in generating continued growth in 12 Subsys usage include taking market share from other 13 competing TIRF products and expanding the usage of Subsys for BTCP by building awareness among oncologists of its 14 rapid onset of action, improved bioavailability, most complete range of dosage strengths and ease of administration relative 15 to other TIRF products. 16 (Doc. 77 at 103-05; see Doc. 107 at 41.) “TIRF” refers to “Transmucosal Immediate- 17 Release Fentanyl.” “BTCP” refers to “breakthrough cancer pain.” 18 Plaintiff also alleges that Insys stock prices declined because of news or 19 announcements to investors that revealed the company’s prior statements to have been false 20 or misleading. On November 4, 2015, CNBC published an article on its website revealing 21 that the Department of Health and Human Services classified Subsys as commonly 22 prescribed for unintended uses. In the context of publicized fraud allegations against Insys 23 employees and collaborating healthcare providers, the article constituted a partial 24 disclosure of the company’s underlying fraud. On December 3, 2015, and January 25, 25 2016, the Southern Investigative Reporting Foundation partially disclosed fraud by Insys 26 in articles about the company’s “Prior Authorization Unit” and Babich’s resignation. 27 28

-3- 1 Plaintiff seeks to certify a class (the “Class”) pursuant to Federal Rules of Civil 2 Procedure 23(a) and 23(b)(3) consisting of all persons and entities who purchased or 3 otherwise acquired Insys common stock during the period from March 3, 2015, through 4 January 25, 2016 (the “Class Period”), and were damaged thereby. Excluded from the 5 Class are Defendants; present and former directors or executive officers of Insys and 6 members of their immediate families; any of the foregoing individuals’ or entities’ legal 7 representatives, heirs, successors, or assigns; and any entity in which any Defendant has or 8 had a controlling interest or which is related to or affiliated with any Defendant. 9 Count I of Plaintiff’s Second Amended Complaint alleges against all Defendants 10 securities fraud in violation of Section 10(b) of the Exchange Act, 15 U.S.C. § 78j(b), and 11 Rule 10b-5 promulgated thereunder by the SEC, 17 C.F.R. § 240.10b-5. Count II alleges 12 control-person liability for Defendants Babich, Baker, and Kapoor under Section 20(a) of 13 the Exchange Act. (See Docs. 77, 107.) 14 II. DEFENDANTS’ MOTION FOR LEAVE TO FILE SUR-REPLY 15 Defendants move for leave to file a sur-reply to respond to arguments and evidence 16 included for the first time in Plaintiff’s reply in support of his motion for class certification. 17 (Doc. 170.) Defendants respond not only to “new” arguments and evidence included in 18 Plaintiff’s reply, but also to issues that were “not addressed in any meaningful detail” in 19 Plaintiff’s opening brief. Plaintiff contends that each of the arguments Defendants describe 20 as “new” is a response to arguments or concessions Defendants did not make until they 21 filed their response to the class certification motion or to Defendants’ attempt to rebut the 22 presumption that proposed class members relied on Defendants’ misstatements. 23 Defendants attached their proposed sur-reply to be filed if leave were granted. Plaintiff’s 24 opposition brief responds both to the motion and the content of the proposed sur-reply. 25 Where new evidence is presented in a reply brief, the district court should not 26 consider the new evidence without giving the non-movant an opportunity to respond. 27 Provenz v. Miller, 102 F.3d 1478, 1483 (9th Cir. 1996). Here, both sides have addressed 28 any “new” evidence and arguments presented in Plaintiff’s reply brief. Because Plaintiff

-4- 1 has responded to the content of Defendants’ proposed sur-reply, there is no prejudice or 2 unfairness in allowing Defendants’ proposed sur-reply to be filed. In deciding Plaintiff’s 3 motion for class certification, the Court considers both Defendants’ sur-reply and 4 Plaintiff’s opposition brief along with all of the other related submissions. 5 III. RULE 23(a) 6 “A party seeking class certification must affirmatively demonstrate his compliance 7 with [Federal Rule of Civil Procedure 23.]” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 8 350 (2011). The party must prove compliance with Rule 23 in fact. Id. “[C]ertification is 9 proper only if the trial court is satisfied, after a rigorous analysis, that the prerequisites of 10 Rule 23(a) have been satisfied.” Id. (internal quotation marks omitted). Although the class 11 certification analysis might include some overlap with the merits of the plaintiff’s 12 underlying claim, merits questions may be considered only to the extent that they are 13 relevant to determining whether the Rule 23 prerequisites for class certification are 14 satisfied. Id. at 351; Amgen Inc. v. Connecticut Ret. Plans & Trust Funds, 568 U.S. 455, 15 466 (2013). 16 Under Rule 23(a), the party seeking certification must prove: 17 (1) the class is so numerous that joinder of all members is impracticable; 18 (2) there are questions of law or fact common to the class; 19 (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and 20 (4) the representative parties will fairly and adequately protect the 21 interests of the class.

22 Defendants do not dispute that the proposed class satisfies Rule 23(a).

23 Numerosity. Rule 23(a)’s numerosity requirement is generally assumed to have

24 been met in class action suits involving nationally traded securities. Zeidman v. J. Ray

25 McDermott & Co., 651 F.2d 1030, 1039 (5th Cir. 1981); see, e.g., In re Infineon Techs. AG

26 Sec. Litig., 266 F.R.D. 386, 393 (N.D. Cal. 2009). “Joinder impracticability (numerosity)

27 is rarely contested in class actions brought on behalf of shareholders or traders in publicly

28 owned corporations. In class actions brought on behalf of securities traders, federal trial

-5- 1 courts are quite willing to accept common sense assumptions in order to support a finding 2 of numerosity.” Zeidman, 651 F.2d at 1039 (quoting 5 J. Newburg, Class Actions § 8812, 3 at 836 (1977)). 4 Plaintiff has submitted evidence, which Defendants do not dispute, that during the 5 Class Period there were between 70 and 72 million shares of Insys common stock 6 outstanding, an average of 4.5 million shares traded weekly, and at least 280 institutional 7 investors held Insys common stock. This satisfies Rule 23(a)(1). 8 Commonality. To demonstrate “questions of law or fact common to the class,” 9 plaintiffs’ claims must depend on a common contention capable of classwide resolution 10 such that determination of the truth of the contention will resolve an issue central to the 11 validity of each of the claims. Wal-Mart, 564 U.S. at 350. Plaintiff has identified numerous 12 common questions of law and fact, which include whether Defendants’ statements 13 misrepresented or omitted facts, whether Defendants’ statements created and/or maintained 14 artificial inflation in the price of Insys common stock, and whether alleged stock price 15 declines are causally connected to Defendants’ misrepresentations and omissions. This 16 satisfies Rule 23(a)(2). 17 Typicality. The typicality requirement ensures that the interest of the class 18 representative aligns with the interests of the class. Just Film, Inc. v. Buono, 847 F.3d 19 1108, 1116 (9th Cir. 2017). Indicators of typicality include whether other members have 20 the same or similar injury, whether the action is based on conduct that is not unique to the 21 named plaintiffs, and whether other class members have been injured by the same course 22 of conduct. Id. “Under the rule’s permissive standards, representative claims are ‘typical’ 23 if they are reasonably co-extensive with those of absent class members; they need not be 24 substantially identical.” Hanlon v. Chrysler Corp., 150 F.3d 1011, 1020 (9th Cir. 1998). 25 The proposed class consists of all persons and entities who purchased or otherwise 26 acquired Insys common stock during the Class Period and were damaged thereby. Plaintiff, 27 like the other members of the proposed class, purchased Insys common stock during the 28 Class Period at prices that Plaintiff alleges were artificially inflated by Defendants’

-6- 1 materially false or misleading statements and allegedly suffered damages when the 2 concealed truth was revealed, causing the stock price to decline. This satisfies Rule 3 23(a)(3). 4 Adequacy of Plaintiff and Class Counsel. The final requirement for class 5 certification under Rule 23(a) is that “the representative parties will fairly and adequately 6 protect the interests of the class.” “Resolution of two questions determines legal adequacy: 7 (1) do the named plaintiffs and their counsel have any conflicts of interest with other class 8 members and (2) will the named plaintiffs and their counsel prosecute the action vigorously 9 on behalf of the class?” Hanlon, 150 F.3d at 1020. 10 Plaintiff has certified that he is willing to serve as a representative party on behalf 11 of the class, including providing testimony at deposition and trial, if necessary. (Doc. 12 34-1.) He further certified that he will not accept any payment for serving as a 13 representative party on behalf of the class beyond his pro rata share of any recovery, except 14 directly related reasonable costs and expenses as ordered by the Court. He disclosed his 15 Class Period purchase and sale transactions in Insys securities that are the subject of this 16 action. Plaintiff submitted information regarding the qualifications of Kessler Topaz 17 Meltzer & Check, LLC, to serve as Class Counsel and those of Bonnett, Fairbourn, 18 Friedman & Balint, P.C., to serve as Liaison Counsel under Rule 23(g). Defendants do not 19 dispute that Plaintiff and his counsel will fairly and adequately protect the interests of the 20 class. 21 Defendants do not dispute that the proposed class satisfies Rule 23(a). 22 IV. RULE 23(b) 23 In addition to satisfying Rule 23(a), the proposed class must satisfy at least one of 24 the three requirements stated in Rule 23(b). Plaintiff relies on Rule 23(b)(3), which 25 requires findings regarding both predominance and superiority: 26 The court finds that the questions of law or fact common to class members predominate over any questions affecting only individual members, and that 27 a class action is superior to other available methods for fairly and efficiently 28 adjudicating the controversy. The matters pertinent to these findings include:

-7- (A) the class members’ interests in individually controlling the 1 prosecution or defense of separate actions; 2 (B) the extent and nature of any litigation concerning the controversy 3 already begun by or against class members;

4 (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and 5 (D) the likely difficulties in managing a class action. 6 Defendants do not dispute that Plaintiff has shown that a class action is superior to other 7 available methods for fairly and efficiently adjudicating the controversy presented in this 8 case. 9 The only remaining decision is whether “the questions of law or fact common to 10 class members predominate over any questions affecting only individual members.” 11 Considering whether “questions of law or fact common to class members 12 predominate” begins, of course, with the elements of the underlying cause of action. The elements of a private securities fraud claim based on violations 13 of § 10(b) and Rule 10b-5 are: (1) a material misrepresentation or omission 14 by the defendant; (2) scienter; (3) a connection between the misrepresentation or omission and the purchase or sale of a security; (4) 15 reliance upon the misrepresentation or omission; (5) economic loss; and (6) 16 loss causation.

17 Erica P. John Fund, Inc. v. Halliburton Co. (“Halliburton I”), 563 U.S. 804, 809-10 (2011)

18 (internal quotation marks and citations omitted). Defendants contend that Plaintiff cannot

19 satisfy Rule 23(b)’s predominance requirement with respect to the questions of reliance

20 and damages.

21 Although reliance by the plaintiff on the defendant’s deceptive acts is an essential

22 element of the § 10(b) private cause of action, limiting proof of reliance to showing that

23 the plaintiff was aware of a company’s statement and engaged in a relevant transaction

24 based on that specific misrepresentation “would place an unnecessarily unrealistic

25 evidentiary burden on the Rule 10b-5 plaintiff who has traded on an impersonal market.”

26 Id. at 810 (quoting Basic Inc. v. Levinson, 485 U.S. 224, 245 (1988)). Moreover, requiring

27 proof of individualized reliance would effectively prevent plaintiffs from proceeding with

28 a class action because individual issues would overwhelm the common ones. Id.

-8- 1 Therefore, in Basic, the Court permitted plaintiffs to invoke a rebuttable presumption of 2 reliance based on the “fraud-on-the-market” theory. Id. 3 A. The Fraud-on-the-Market Rebuttable Presumption of Reliance 4 The Basic rebuttable presumption of reliance is based on the theory that the market 5 price of shares traded on well-developed markets reflects all publicly available 6 information, including material misrepresentations. Id. at 811. It is assumed that an 7 investor relies on public misstatements whenever he buys or sells stock at market prices. 8 Id. 9 This presumption springs from the very concept of market efficiency. If a market is generally efficient in incorporating publicly available information 10 into a security’s market price, it is reasonable to presume that a particular 11 public, material misrepresentation will be reflected in the security’s price. Furthermore, it is reasonable to presume that most investors—knowing that 12 they have little hope of outperforming the market in the long run based solely 13 on their analysis of publicly available information—will rely on the security’s market price as an unbiased assessment of the security’s value in 14 light of all public information. Thus, courts may presume that investors 15 trading in efficient markets indirectly rely on public, material misrepresentations through their reliance on the integrity of the price set by 16 the market. 17 Amgen Inc. v. Connecticut Ret. Plans & Trust Funds, 568 U.S. 455, 462 (2013). The fraud- 18 on-the-market presumption can be rebutted by appropriate evidence. Halliburton I, 563 19 U.S. at 811. “Any showing that severs the link between the alleged misrepresentation and 20 either the price received (or paid) by the plaintiff, or his decision to trade at a fair market 21 price, will be sufficient to rebut the presumption of reliance.” Basic, 485 U.S. at 248. 22 However, 23 Basic’s presumption of reliance . . . does not rest on a ‘binary’ view of market efficiency. Indeed, in making the presumption rebuttable, Basic recognized 24 that market efficiency is a matter of degree and accordingly made it a matter 25 of proof. 26 . . . Even the foremost critics of the efficient-capital-markets hypothesis acknowledge that public information generally affects stock prices. . . . 27 Debates about the precise degree to which stock prices accurately reflect 28 public information are thus largely beside the point.

-9- 1 Erica P. John Fund, Inc. v. Halliburton Co. (“Halliburton II”), 573 U.S. 258, 272 (2014). 2 Basic’s fraud-on-the-market presumption does not relieve plaintiffs of the burden 3 of proving, not simply pleading, the predominance requirement of Rule 23(b)(3) before 4 class certification. Id. at 276. To demonstrate that the presumption of reliance applies, a 5 plaintiff must prove: “(1) the alleged misrepresentations were publicly known, (2) they 6 were material, (3) the stock traded in an efficient market, and (4) the plaintiff traded the 7 stock between when the misrepresentations were made and when the truth was revealed.” 8 Id. at 277-78. The first three prerequisites are directed at whether the alleged 9 misrepresentations affected the market price, i.e., price impact. Id. at 278. Plaintiffs are 10 not required to prove price impact directly; such a requirement would eliminate the 11 presumption that a misrepresentation affected the stock price. Id. at 279. 12 Defendants may defeat the fraud-on-the-market presumption of reliance at the class 13 certification stage with evidence that the misrepresentation in fact did not affect the stock 14 price. Id. Defendants may introduce price impact evidence at the class certification stage, 15 but only “for the purpose of countering a plaintiff’s showing of market efficiency, rather 16 than directly rebutting the presumption.” Id. at 280. While Basic allows plaintiffs to 17 establish price impact indirectly, “it does not require courts to ignore a defendant’s direct, 18 more salient evidence showing that the alleged misrepresentation did not actually affect 19 the stock market’s price and, consequently, that the Basic presumption does not apply.” 20 Id. at 282. 21 Although materiality is an essential predicate of the fraud-on-the-market theory, 22 Rule 23(b) does not require a plaintiff invoking the fraud-on-the-market presumption of 23 reliance to prove materiality at the class certification stage because such proof is not 24 necessary to ensure satisfaction of the predominance requirement. Amgen, 568 U.S. at 466, 25 467 & n.4. Materiality is a common question for purposes of Rule 23(b)(3), and failure of 26 proof on that common question will not result in individual questions predominating. Id. 27 at 467-68. 28

- 10 - 1. Factors that Demonstrate Market Efficiency 1 Defendants contend that Plaintiff is not entitled to presumption of reliance based on 2 the fraud-on-the-market theory because he has not shown that the market is efficient. To 3 determine whether a market is efficient, i.e., whether stock prices reflect public 4 information, the Ninth Circuit and many other courts have considered five characteristics 5 of the company and its stock (referred to as the Cammer factors): (1) whether the stock 6 trades at a high weekly volume; (2) whether securities analysts follow and report on the 7 stock; (3) whether the stock has market makers and arbitrageurs; (4) whether the company 8 is eligible to file SEC registration form S-3; and (5) whether there are “empirical facts 9 showing a cause and effect relationship between unexpected corporate events or financial 10 releases and an immediate response in the stock price.” Binder v. Gillespie, 184 F.3d 1059, 11 1065 (9th Cir. 1999) (quoting Cammer v. Bloom, 711 F. Supp. 1264, 1286-87 (D.N.J. 12 1989)). The fifth characteristic, empirical evidence of a causal relationship, “is the essence 13 of an efficient market and the foundation for the fraud on the market theory.” Cammer, 14 711 F. Supp. at 1287. “[I]n an efficient market, a stock’s price remains relatively stable in 15 the absence of news, and changes very rapidly as the market receives new and unexpected 16 information.” Krogman v. Sterritt, 202 F.R.D. 467, 474 (N.D. Tex. 2001). 17 Here, the parties dispute whether the fraud-on-the-market presumption requires 18 plaintiffs to prove satisfaction of the fifth Cammer factor. Defendants do not dispute that 19 the first four Cammer factors are satisfied here, but they contend that Plaintiff must meet, 20 and has not met, his burden of proving the fifth Cammer factor. Plaintiff contends that he 21 has done so and also that it is not necessary to prove the fifth Cammer factor where the 22 first four factors are satisfied and additional factors demonstrate market efficiency. 23 The Cammer court stated that it is not logical to draw bright line tests to assist fact 24 finders in determining whether a stock trades in an open and efficient market. Cammer, 25 711 F. Supp. at 1287. The Ninth Circuit described the Cammer factors as “designed to 26 help make the central determination of efficiency in a particular market.” Binder, 184 F.3d 27 at 1065. In a different context, the Ninth Circuit described the Cammer factors as setting 28

- 11 - 1 a necessarily “high bar” because “[d]ue process concerns require that class certification 2 meet rigorous standards in securities cases.” Miller v. Thane Int’l, Inc., 615 F.3d 1095, 3 1103 (9th Cir. 2010). However, the only specific guidance from the Ninth Circuit is that 4 evidence supporting only the third factor, the existence of market makers and arbitrageurs, 5 is insufficient to invoke the presumption of reliance. Binder, 184 F.3d at 1065. 6 Some courts have held that the Cammer factors can be supplemented by other 7 measures, such as the company’s market capitalization, the bid-ask spread, the issue 8 amount outstanding excluding insider-owned securities, and the percentage of institutional 9 ownership. Petrie v. Electronic Game Card, Inc., 308 F.R.D. 336, 349 (C.D. Cal. 2015). 10 Three of these factors are commonly referred to as the “Krogman factors”: “(1) the 11 capitalization of the company; (2) the bid-ask spread of the stock; and (3) the percentage 12 of stock not held by insiders (the ‘float’’).” Krogman v. Sterritt, 202 F.R.D. 467, 474 (N.D. 13 Tex. 2001). In Krogman, the court found two of the five Cammer factors weighed in favor 14 of market efficiency, and two weighed against; regarding the fifth factor, the relationship 15 of price changes to news was “inconsistent at best.” Id. at 477. Of the three additional 16 factors, one weighed slightly in favor of a determination of market efficiency, and two did 17 not. Id. at 478. The court concluded that the presumption of reliance did not apply. 18 The Second, Fourth, Fifth, and Eleventh Circuits have instructed that the Cammer 19 factors serve only as a guide for determining market efficiency to be applied in a case-by- 20 case basis in addition to other considerations and that, while important in certain cases, the 21 fifth Cammer factor is not a mandatory prerequisite in every case for finding market 22 efficiency. Waggoner v. Barclays PLC, 875 F.3d 79, 97-98 (2d Cir. 2017); Gariety v. 23 Grant Thornton, LLP, 368 F.3d 356, 368 (4th Cir. 2004); Unger v. Amedisys Inc., 401 F.3d 24 316, 323 (5th Cir. 2005); Bell v. Ascendant Solutions, Inc., 422 F.3d 307, 313, 316 (5th 25 Cir. 2005); Local 703, I.B. of T. Grocery & Food Employees Welfare Fund v. Regions 26 Financial Corp., 762 F.3d 1248, 1255-56 (11th Cir. 2014). Although these circuit courts 27 require proof and thorough analysis of market efficiency in context, they disclaim rigid 28

- 12 - 1 adherence to a bright line test. The parties have not cited, and the Court has not found, 2 contrary authority. 3 In Regions, the Eleventh Circuit concluded that it was not wise to require district 4 courts to analyze market efficiency in terms of the Cammer factors in every case and better 5 to allow district courts flexibility to perform the analysis “in the most sensible way given 6 the circumstances.” 762 F.3d at 1256. Specifically, the Eleventh Circuit stated that a 7 finding of market efficiency does not always require proof that the alleged 8 misrepresentations had an immediate effect on the stock price. For example, when a 9 company releases expected information, regardless of whether it is truthful, the theory 10 underlying the fraud-on-the-market presumption predicts that the disclosure will cause no 11 significant change in the stock price. Further, 12 Neither would it make sense to impose an unwavering requirement for plaintiffs to identify unexpected disclosures during or around the class period 13 that had an immediate price impact. In any given case there may be no 14 unexpected disclosures during the period at all, because the company is withholding that information. To require plaintiffs to prove a set number of 15 unexpected disclosures resulting in an immediate price impact would rob 16 District Courts of the flexibility they need to conduct holistic, fact-sensitive inquiries into the efficiency of the market for the particular stock before it. 17 Id. at 1257. In Regions, the Eleventh Circuit affirmed the district court’s finding that the 18 presumption of classwide reliance was justified and remanded for the district court to 19 determine whether the presumption had been rebutted. 20 Similarly, the Second Circuit said that the fifth Cammer factor might be the most 21 important factor is certain cases, such as when the other Cammer and/or Krogman factors 22 are less compelling, but a plaintiff need not always present direct evidence of price impact 23 through event studies to demonstrate market efficiency. Waggoner, 875 F.3d at 97-98. In 24 Waggoner, the first four Cammer factors and the three Krogman factors weighed so clearly 25 in favor of concluding that the market was efficient that the defendants did not even 26 challenge them. Thus, the district court’s decision not to rely on direct evidence of price 27 impact under the fifth Cammer factor fell within its discretion. Id. at 98. 28

- 13 - 1 In Unger, the Fifth Circuit said the five Cammer factors and the three Krogman 2 factors do not represent an exhaustive list of factors to be considered in determining market 3 efficiency, and in some cases one of those factors might be unnecessary. 401 F.3d at 323. 4 But the factors “must be weighed analytically, not merely counted, as each of them 5 represents a distinct facet of market efficiency.” Id. The Fifth Circuit held that the district 6 court devoted insufficient attention to evaluating the market efficiency factors. The district 7 court had found a high stock trading volume, but the plaintiffs never proved the actual 8 number of shares being regularly traded. The district court had relied on the number of 9 market makers with no further analysis. And the district court had found a causal 10 connection between corporate events and movement of the stock price, but it did not take 11 into account the many other factors that could affect the price of the stock. Moreover, the 12 district court failed to evaluate the significance of the factors that were lacking. 13 In Bell, the Fifth Circuit affirmed both the district court’s exclusion of expert 14 testimony regarding market efficiency found to be unreliable and its denial of class 15 certification. 422 F.3d at 316. The plaintiffs did not provide any analysis of market 16 efficiency factors except for the excluded expert testimony, and there was insufficient 17 evidence to support finding the market was efficient. The appellate court found no abuse 18 of discretion in denying class certification where the plaintiffs’ expert report was excluded, 19 the plaintiffs’ briefing to the district court was “devoid of any serious effort to show market 20 efficiency,” and the plaintiffs bore the burden of proof. Id. 21 In summary, Plaintiff bears the burden of proving market efficiency required for 22 application of the fraud-on-the-market presumption of reliance. There is no bright line test 23 for market efficiency. The Court may consider evidence related to the Cammer factors, 24 the Krogman factors, and/or any other market characteristics relevant to a specific case. 25 The factors provide a framework for thorough analysis, not a checklist for keeping score. 26 Evidence of the market’s reaction to unexpected corporate events or financial disclosures 27 is usually important, but not required in every case. 28

- 14 - 2. Evidence Regarding the Fifth Cammer Factor 1 Defendants do not dispute that Plaintiff has shown sufficient evidence of the first 2 four Cammer factors. They contend only that Plaintiff has not established the fifth factor, 3 which is whether there are “empirical facts showing a cause and effect relationship between 4 unexpected corporate events or financial releases and an immediate response in the stock 5 price.” Cammer, 711 F. Supp. at 1287. 6 Plaintiff has submitted evidence that during the Class Period Insys common stock 7 had a high weekly trading volume, significant analyst coverage, and 97 market makers. 8 Plaintiff’s expert, Chad Coffman, CFA, found no evidence that Insys was not eligible to 9 file a SEC registration form S-3. To satisfy the fifth Cammer factor, Plaintiff relies on an 10 event study conducted by Coffman, which Plaintiff contends provides direct evidence that 11 the price of Insys common stock promptly responded to the release of Insys-specific 12 information during the Class Period. 13 As described by Coffman, “[a]n event study is a technique used to measure the effect 14 of new information on the market prices of a company’s publicly traded securities.” (Doc. 15 160-3 at 21.) “An event study is conducted by specifying a model of expected price 16 movements conditioned on outside market factors and then testing whether the deviation 17 from expected price movements is sufficiently large that simple random movement can be 18 rejected as the cause.” Id. Coffman’s event study compared Insys’s stock price movements 19 on 15 days when Insys released its earnings results (“news days”) to days when no news 20 regarding Insys was released (“non-news days”). Coffman concluded that the Insys 21 common stock price was far more likely to exhibit statistically significant price reaction on 22 “news days” than on “non-news days.” In Coffman’s opinion, Insys common stock 23 experienced a statistically significant price reaction at the 95% confidence level on 33.33% 24 of the “news days,” compared to 2.60% of “non-news days.” 25 The Class Period is March 3, 2015, through January 25, 2016. Coffman determined 26 that the Class Period consisting of 227 trading days was “not sufficiently long to provide 27 enough data with which to test cause-and-effect in isolation.” (Doc. 160-3 at 25 n.61.) 28

- 15 - 1 Therefore, Coffman examined the price response of Insys common stock to 15 earnings 2 announcements that occurred during the “Analysis Period,” which Coffman defined as four 3 quarters before the Class Period through four quarters after the Class Period. (Id. at 25.) 4 Only four of the 15 earnings announcements Coffman examined were issued during the 5 Class Period. (Id.) Of the 15 earnings announcements issued during the Analysis Period, 6 five resulted in statistically significant price movements above the 95% confidence level; 7 only one of the five was issued during the Class Period. Coffman concluded that on five 8 of the 15 (33.33%) “news days” Insys common stock experienced a statistically significant 9 price movement. Coffman then examined the 154 “non-news days” during the Analysis 10 Period and found four statistically significant price movements, which comprised 2.60% 11 of the “non-news days.” He concluded that 33.33% is statistically significantly different 12 from 2.60%. 13 To establish a presumption of reliance, Plaintiff must show that the market was 14 efficient during the period for which the presumption of reliance is sought, i.e., the Class 15 Period. Plaintiff offers no basis for assuming that the market’s responsiveness a year before 16 and a year after the Class Period is representative of the market’s responsiveness during 17 the Class Period. A longer period for analysis is necessary only because there are too few 18 events during the Class Period to draw statistical conclusions. During the Class Period, 19 there were only four “news days,” and only one of those was associated with a statistically 20 significant price reaction. However, even considering the entire Analysis Period, only one- 21 third of the “news days” were associated with statistically significant price movements; 22 two-thirds were not. Moreover, during the Analysis Period, there were nine days during 23 which Insys common stock experienced a statistically significant price movement, five of 24 which occurred on “news days” and four of which occurred on “non-news days.” And 25 there were ten “news days” during which Insys common stock did not experience a 26 statistically significant price movement. There are just too few “news days” to observe a 27 meaningful correlation between “news” and “price reaction.” 28

- 16 - 1 Further, Coffman characterized each of the days on which an earnings report was 2 released as a “news day” regardless of whether the “news” was expected or unexpected, 3 and he did not report regarding any “news” other than earnings reports. The fifth Cammer 4 factor considers whether there is a correlation between unexpected public information and 5 a change in a stock’s price; information that is consistent with expectations likely would 6 not affect a stock’s price. Moreover, finding an association between “news” and “price 7 reaction” does not demonstrate that “good news” correlated with price increases and “bad 8 news” correlated with price decreases. 9 Defendants’ rebuttal expert, David C. Smith, Ph.D., was tasked with assessing 10 whether Coffman had reliably demonstrated that the market for Insys common stock was 11 efficient during the Class Period. Smith opined that finding market efficiency requires 12 satisfaction of the fifth Cammer factor, regardless of evidence regarding other factors, and 13 that flaws in Coffman’s methodology renders his analysis incapable of reliably assessing 14 market efficiency of Insys common stock during the Class Period. Among other things, 15 Smith contends that an efficient market responds rapidly to new, value-relevant 16 information nearly 100% of the time and Coffman failed to assess the speed, size, and 17 direction of the stock price response to “news.” Smith further contends Coffman failed to 18 establish that what he characterized as “news” was in fact new, value-relevant information. 19 Smith provides examples of varying and inconsistent returns to similarly positive 20 unexpected information and similarly positive analyst commentary, demonstrating market 21 inefficiency, that Coffman has not explained. Coffman responds to each of Smith’s 22 criticisms and provides additional statistical analysis, but fails to overcome the fact that 23 there were too few “news days” during the Class Period or the Analysis Period for 24 statistical significance to have persuasive meaningful significance. 25 3. Additional Factors 26 Plaintiff provided evidence, which Defendants do not dispute, that additional factors

27 support finding market efficiency in this case. Coffman, Plaintiff’s expert, reported that

- 17 - 1 Insys’s average market capitalization ($2.2 billion) was greater than at least 68% of the 2 companies whose securities traded on the NYSE and NASDAQ during the Class Period. 3 Regarding the bid-ask spread, Coffman explained that the narrower the bid-ask 4 spread, the greater the indication of an efficient market. Coffman determined that the 5 average bid-ask spread for Insys common stock during the Class Period was between 6 0.12% and 0.36%, which was well below the average of 0.97% of a randomly selected 7 group of 100 other common stocks on the NASDAQ. 8 Coffman reported that 280 institutions reported owning Insys common stock during 9 the Class Period, holding on average 87.2% of public float. He opined that this substantial 10 level of institutional ownership, coupled with the high trading volume, supports a 11 conclusion of market efficiency. 12 Autocorrelation measures the relationship between a variable’s current value and its 13 past values. If past price movements of a security predict future price movements, the 14 prices are said to be “autocorrelated.” If autocorrelation is persistent and sufficiently large, 15 a trader can consistently earn riskless profits over time, and the market is considered 16 inefficient. Coffman’s statistical analysis showed there was no evidence of autocorrelation 17 for Insys common stock during the Class Period. 18 Finally, Coffman found considerable option trading in Insys common stock during 19 the Class Period, which in Coffman’s opinion supports finding that Insys common stock 20 traded in an efficient market throughout the Class Period. 21 4. Market Efficiency Findings 22 As explained above, there is no bright line test for determining market efficiency.

23 Multiple factors are to be considered and do not necessarily deserve equal weight.

24 Plaintiffs may prove market efficiency without satisfying the fifth Cammer factor. See,

25 e.g., Angley v. UTi Worldwide Inc., 311 F. Supp. 3d 1117, 1121 (C.D. Cal. 2018) (district

26 court concluded it was not necessary to reach a conclusion regarding the fifth Cammer

27 factor); Petrie v. Electronic Game Card, Inc., 308 F.R.D. 336, 358 (C.D. Cal. 2015)

28 (district court found that the plaintiffs had met their burden of proving that they were

- 18 - 1 entitled to the fraud-on-the-market presumption where plaintiffs had “relatively weak 2 evidence of market efficiency” and “the most important cause-and-effect factor” weighed 3 only weakly in favor of the plaintiffs); Smilovits v. First Solar, Inc., 295 F.R.D. 423, 437 4 (D. Ariz. 2013) (district court found that the plaintiffs had shown the stock traded on an 5 efficient market by showing that three of the five Cammer factors were satisfied, another 6 was partially satisfied, and the stock traded on the NASDAQ even though the evidence was 7 “essentially a stand-off on the fifth Cammer factor”). 8 Here, Defendants do not dispute that four of the five Cammer factors, all of the 9 Krogman factors, and several other factors are satisfied. Although the fifth Cammer factor 10 is important, the scarcity of “news days” during the relevant period precludes meaningful 11 statistical analysis. Concluding that under these circumstances it is not necessary to satisfy 12 the fifth Cammer factor, the Court finds that Insys common stock traded in an efficient 13 market throughout the Class Period. 14 Therefore, under Basic’s fraud-on-the-market theory, Plaintiff is entitled to a 15 presumption of reliance based on indirect evidence of price impact unless Defendants 16 produce “direct, more salient evidence showing that the alleged misrepresentation did not 17 actually affect the stock’s market price and, consequently, that the Basic presumption does 18 not apply.” Halliburton II, 573 U.S. at 282. 19 B. Defendants’ Rebuttal of the Basic Presumption 20 Although plaintiffs do not need to directly prove price impact to invoke the Basic 21 presumption, defendants may defeat the presumption at the class certification stage through 22 direct evidence that the misrepresentation at issue did not in fact affect the stock price. Id. 23 at 279, 282. “Any showing that severs the link between the alleged misrepresentation and 24 either the price received (or paid) by the plaintiff, or his decision to trade at a fair market 25 price, will be sufficient to rebut the presumption of reliance.” Basic, 485 U.S. at 248. “The 26 Supreme Court provided a non-exhaustive list of ways that defendants can rebut the 27 presumption, including by showing: (1) the market did not respond to the alleged 28 misrepresentations; (2) the misrepresentations were immaterial; (3) a plaintiff did not

- 19 - 1 actually rely on the misrepresentations; or (4) a plaintiff would have sold the securities 2 without relying on the integrity of the market.” In re DVI, Inc. Sec. Litig., 639 F.3d 623, 3 637 (3d Cir. 2011) (summarizing Basic, 485 U.S. at 248-49). 4 1. Burden of Proof for Rebuttal 5 The parties dispute whether in order to successfully rebut the Basic presumption

6 Defendants must demonstrate a lack of price impact by a preponderance of the evidence or

7 merely meet a burden of production. Neither the Supreme Court nor the Ninth Circuit has

8 expressly decided the question.

9 Federal Rule of Evidence 301 provides:

10 In a civil case, unless a federal statute or these rules provide otherwise, the party against whom a presumption is directed has the burden of producing 11 evidence to rebut the presumption. But this rule does not shift the burden of 12 persuasion, which remains on the party who had it originally.

13 The Advisory Committee’s Notes to Rule 301 (1974 Enactment) explain that the

14 Committee adopted “an intermediate position under which a presumption does not vanish

15 upon the introduction of contradicting evidence, and does not change the burden of

16 persuasion; instead it is merely deemed sufficient evidence of the fact presumed, to be

17 considered by the jury or other finder of fact.” The Notes further state: If the adverse party offers no evidence contradicting the presumed fact, the 18 court will instruct the jury that if it finds the basic facts, it may presume the 19 existence of the presumed fact. If the adverse party does offer evidence contradicting the presumed fact, the court cannot instruct the jury that it may 20 presume the existence of the presumed fact from proof of the basic facts. The 21 court may, however, instruct the jury that it may infer the existence of the presumed fact from proof of the basic facts. 22 Applying Rule 301, therefore, if persuaded, the court as trier of fact may infer the existence 23 of the presumed fact (i.e., the Basic presumption of reliance) upon proof of the prerequisites 24 for the Basic presumption even though the adverse party has produced contradicting 25 evidence. 26 The Second Circuit has held that “defendants must rebut the Basic presumption by 27 disproving reliance by a preponderance of the evidence at the class certification stage.” 28

- 20 - 1 Waggoner v. Barclays PLC, 875 F.3d 79, 99 (2d Cir. 2017); accord Arkansas Teachers 2 Ret. Sys. v. Goldman Sachs Group, Inc., 879 F.3d 474, 485 (2d Cir. 2018). In Waggoner, 3 the Second Circuit rejected the defendants’ contention that Federal Rule of Evidence 301 4 imposed a burden of production, not persuasion, for rebutting the Basic presumption. 875 5 F.3d at 102-03. It reasoned that because the Basic presumption was adopted by the 6 Supreme Court pursuant to federal securities laws and has been described by the Court as 7 a substantive doctrine of federal securities fraud law, there is a sufficient link to satisfy 8 Rule 301’s exception for federal statutory provisions. The Second Circuit did not explain 9 exactly how federal securities law provides a rule other than Rule 301. 10 In Waggoner, the court did explain that it would be inconsistent with Halliburton II 11 to require plaintiffs to prove satisfaction of the predominance requirement, including the 12 prerequisites for invoking the Basic presumption of reliance, “while allowing defendants 13 to rebut the Basic presumption by simply producing some evidence of market inefficiency, 14 but not demonstrating its inefficiency to the district court. The presumption of reliance 15 would also be of little value if it were so easily overcome.” 875 F.3d at 100-01. The court 16 further found that the specific language of Halliburton II “demonstrates that the Court 17 understood the burden that shifts to defendants as one of persuasion rather than 18 production”: 19 [T]he majority in Halliburton II explained that evidence that satisfied the “severing the link” standard would rebut the Basic presumption because 20 “‘the basis for finding that the fraud had been transmitted through market 21 price would be gone,’” and the defendants’ “direct, more salient evidence” that the misrepresentations did not affect the stock price would rebut the 22 Basic presumption. 23 Waggoner, 875 F.3d at 101-02 (quoting Halliburton II, which quotes Basic). “If a 24 defendant shows that an alleged misrepresentation did not, for whatever reason, actually 25 affect the market price of defendant’s stock, there is no grounding for any contention that 26 the investor indirectly relied on that misrepresentation through his reliance on the integrity 27 of the market price.” Arkansas Teachers, 879 F.3d at 486. 28

- 21 - 1 In Waggoner, the Second Circuit also rejected the defendants’ reliance on IBEW 2 Local 98 Pension Fund v. Best Buy Co., 818 F.3d 775 (8th Cir. 2016). 875 F.3d at 103 3 n.36. In Best Buy, the Eighth Circuit quoted Rule 301 referring to “the burden of producing 4 evidence to rebut the presumption” and stated, “[W]hen plaintiffs presented a prima facie 5 case that the Basic presumption applies to their claims, defendants had the burden to come 6 forward with evidence showing a lack of price impact.” 818 F.3d at 782. The Second 7 Circuit did not read Best Buy as deciding the extent of the burden of proof because it was 8 not at issue and the Eighth Circuit described the rebuttal evidence as “overwhelming.” 9 Waggoner, 875 F.3d at 103 n.36. 10 Although Rule 301 provides an exception where “a federal statute or these rules 11 provide otherwise,” it is not necessary here to find that the Basic presumption of reliance 12 is a substantive doctrine of federal securities fraud law that requires shifting the burden of 13 persuasion to the party opposing the presumption. In this case, applying Rule 301’s 14 “intermediate position,” the Court can infer the existence of the presumed fact (i.e., 15 reliance) from Plaintiff’s proof based on indirect evidence of price impact because 16 Defendants have not made “[a]ny showing that severs the link between the alleged 17 misrepresentation and either the price received (or paid) by the plaintiff, or his decision to 18 trade at a fair market price.” See Basic, 485 U.S. at 248. 19 2. Rebuttal Evidence 20 Defendants contend they have rebutted any presumption of reliance “through direct,

21 empirical evidence” because they have used Plaintiff’s expert report to demonstrate that

22 the two alleged misrepresentations at issue had no statistically significant impact. (Doc.

23 165 at 16.) Defendants assert that the two statements were made on March 3, 2015, and

24 that, based on Plaintiff’s expert’s analysis, there was not a statistically significant change

25 in stock price on March 3, 2015, or March 4, 2015. (Id. at 17-18.) However, Plaintiff’s

26 failure to show a price reaction on March 3, 2015, or March 4, 2015, at the 95% confidence

27 level is not direct, empirical evidence at the 95% confidence level of the absence of a price

28 reaction. The lack of statistically significant proof that a statement affected the stock price

- 22 - 1 is not statistically significant proof of the opposite, i.e., that it did not actually affect the 2 stock price. 3 Defense expert Smith testified that he was not offering an opinion that the alleged 4 misrepresentations had no impact on the price of Insys common stock. (Doc. 169-1 at 5 5:18-23.) Rather, Smith opined that Coffman’s analysis showed that there was no 6 statistically significant stock price change on the dates of the two alleged 7 misrepresentations. (Id. at 6:19-25.) 8 Moreover, Coffman’s analysis did not assess the direct price impact of any specific 9 statement. It focused on market efficiency based on earnings reports. Coffman reported 10 that of the 15 earnings announcements Insys issued during the Analysis Period, only five 11 resulted in statistically significant price movements above the 95% confidence level. (Doc. 12 160-3 at 26.) Coffman explained: 13 In any event, even though we observe a number of statistically significant price movements, it is not at all unusual for earnings announcements to be 14 accompanied by price movements that are not statistically significant (after 15 all, requiring 95% confidence that a stock price movement was caused by information sets a high bar for evidence of cause and effect). This would 16 happen, for instance, in quarters where there was not much surprise and the 17 firm roughly met market expectations, if the firm offered little change in guidance, and/or if there was a mix of both positive and negative information 18 that netted out. 19 (Id. at 26 n.65.) 20 Coffman also reported that on March 3, 2015, Insys issued a press release before 21 market hours announcing 4Q 2014 earnings results and a few hours later held a conference 22 call to discuss the earnings announcement. (Doc. 169-2 at 27.) Both the press release and 23 conference call announced positive earnings surprises and negative results, including 24 higher than anticipated legal fees and costs for research and development. (Id. at 27-28.) 25 Market analysts were generally pleased with the company’s performance, but factored the 26 greater than expected costs into their forecasts. (Id. at 28-29.) Coffman’s analysis found 27 that on March 3, 2015, Insys common stock increased by 0.36% after controlling for market 28

- 23 - 1 and industry effects, which is directionally consistent with the total mix of information 2 released and analyst reaction, but is not statistically significant. (Id. at 30.) 3 Defendants have not produced evidence that severs the link between the alleged 4 misrepresentations and the price paid or received by Plaintiff. They have not shown that 5 the market in fact did not respond to the alleged misrepresentations. They have shown 6 only that Plaintiff has not proved that it did, and Plaintiff is not required to do so at the 7 class certification stage. Because Plaintiff has proved the basic facts required by Basic, 8 and Defendants have not rebutted the Basic presumption of reliance, Plaintiff is entitled to 9 a presumption of reliance. 10 C. The Affiliated Ute Presumption of Reliance 11 In addition to the fraud-on-the-market presumption of reliance, reliance also may be 12 presumed in cases “involving primarily a failure to disclose” when the information 13 withheld is “material in the sense that a reasonable investor might have considered them 14 important in the making of this decision.” Affiliated Ute Citizens v. United States, 406 U.S. 15 128, 153-54 (1972) (concluding sellers of stock had a right to know that the defendants 16 who facilitated the sales would gain financially from the sales and that the stock was selling 17 for a higher price in the market). However, 18 [S]ecurities laws do not create an affirmative duty to disclose any and all material information. Instead, companies can control what they have to 19 disclose under these provisions by controlling what they say to the market. 20 But once defendants choose to tout positive information to the market, they are bound to do so in a manner that wouldn’t mislead investors, including 21 disclosing adverse information that cuts against the positive information. 22 Schueneman v. Arena Pharmaceuticals, Inc., 840 F.3d 698, 705-06 (9th Cir. 2016) (quoting 23 Matrixx Initiatives, Inc. v. Siracusano, 563 U.S. 27, 44-45 (2011) and Berson v. Applied 24 Signal Tech., Inc., 527 F.3d 982, 987 (9th Cir. 2008); citations, alterations, and internal 25 quotation marks omitted). 26 The Affiliated Ute presumption of reliance is limited to cases that can be 27 characterized as primarily alleging omissions. Desai v. Deutsche Bank Sec. Ltd., 573 F.3d 28

- 24 - 1 931, 940 (9th Cir. 2009). Omissions “stem from the failure to disclose accurate information 2 relating to the value of a security where one has a duty to disclose it.” Id. 3 Any fraudulent scheme requires some degree of concealment, both of the truth and of the scheme itself. We cannot allow the mere fact of this 4 concealment to transform the alleged malfeasance into an omission rather 5 than an affirmative act. To do otherwise would permit the Affiliated Ute presumption to swallow the reliance requirement almost completely. 6 Moreover, it would fail to serve the Affiliated Ute presumption’s purpose 7 since this is not a case where reliance would be difficult to prove because it was based on a negative. 8 Id. at 941 (quoting Joseph v. Wiles, 223 F.3d 1155, 1163 (10th Cir. 2000)). “[T]he 9 Affiliated Ute presumption should not be applied to cases that allege both misstatements 10 and omissions unless the case can be characterized as one that primarily alleges omissions.” 11 Binder v. Gillespie, 184 F.3d 1059, 1064 (9th Cir. 1999). 12 Plaintiff contends that he is entitled to the Affiliated Ute presumption of reliance “at 13 least with respect to those claims that rely primarily on omissions.” (Doc. 159 at 21.) He 14 does not, however, identify which claims rely primarily on omissions. Nor does he identify 15 what information Defendants had a duty to disclose. 16 The portions of the Second Amended Complaint that were not dismissed (Docs. 77, 17 107) allege that, during a conference call with analysts, Defendant Babich said, “We keep 18 hitting new highs in the number of new doctors that we activate on a weekly basis. We 19 have some very unique programs within the oncology setting that we continue to execute 20 on and any growth that we see in this overall TIRF class is specifically coming from 21 Subsys.” (Doc. 77 at 103; see Doc. 107 at 41.) In addition, the FY14 Form 10-K stated 22 that “some of the key factors in generating continued growth in Subsys usage include 23 taking market share from other competing TIRF products and expanding the usage of 24 Subsys for BTCP by building awareness among oncologists . . . .” (Doc. 77 at 105; see 25 Doc. 107 at 41.)2 The complaint alleges that these statements “were materially false and 26 27 2 “TIRF” refers to “Transmucosal Immediate-Release Fentanyl.” “BTCP” refers to 28 “breakthrough cancer pain.”

- 25 - 1 misleading when made or omitted material information such that they were rendered 2 misleading.” (Doc. 77 at 105.) The complaint further alleges that Defendants knew or 3 were deliberately reckless in disregarding that “new highs in the number of new doctors” 4 and “continued growth in Subsys usage” were the result of illegal off-label promotion, 5 illegal kickbacks to prescribers, and insurance fraud, not from “building awareness among 6 oncologists.” 7 Presumably, reasonable investors would have considered information regarding 8 illegal off-label promotion, illegal kickbacks to prescribers, and insurance fraud to be 9 material. They likely would have considered material the information that Insys sales 10 representatives were instructed to avoid meeting with oncologists and pain specialists and 11 to aggressively pursue prescribers who did not have cancer patients. As alleged, 12 Defendants had a duty to disclose this information because they touted inconsistent positive 13 information to the market. But this case cannot be characterized as one that primarily 14 alleges omissions. Therefore, the Affiliated Ute presumption of reliance does not apply 15 here. 16 D. Plaintiff’s Damages Methodology 17 “To prevail on the merits in a private securities fraud action, investors must 18 demonstrate that the defendant’s deceptive conduct caused their claimed economic loss.” 19 Halliburton I, 563 U.S. at 807. However, they are not required to prove loss causation at 20 the class certification stage. Id. At the class certification stage, the plaintiffs need only 21 present a workable method for calculating damages on a classwide basis that is consistent 22 with their theory of liability. 23 Under Rule 23(b)(3), a plaintiff must show a classwide method for damages 24 calculations as part of the assessment of whether common questions predominate over 25 individual questions. Lambert v. Nutraceutical Corp., 870 F.3d 1170, 1182 (9th Cir. 2017), 26 rev’d on other grounds, 139 S. Ct. 710 (2019). “Uncertainty regarding class members’ 27 damages does not prevent certification of a class as long as a valid method has been 28 proposed for calculating those damages.” Id. A plaintiff must present “a workable

- 26 - 1 method.” Id. at 1184. “Courts regularly reaffirm that the out-of-pocket, or event study, 2 method matches plaintiffs’ theory of liability under Section 10(b) of the Securities 3 Exchange Act, making it the standard method for calculating damages in virtually every 4 Section 10(b) class action.” City of Miami Gen. Employees’ & Sanitation Employees’ Ret. 5 Trust v. RH, Inc., No. 17-CV-00554-YGR, 2018 WL 4931543, at *3 (N.D. Cal. Oct. 11, 6 2018) (collecting cases). 7 “[A]t the class-certification stage (as at trial), any model supporting a plaintiff’s 8 damages case must be consistent with its liability case.” Comcast Corp. v. Behrend, 569 9 U.S. 27, 35 (2013) (internal quotation marks and citation omitted). In Comcast, customers 10 brought an antitrust class action against a cable television company and its subsidiaries 11 based on the company’s strategy of acquiring competitor cable providers in a certain region 12 in exchange for systems outside the region, which allegedly eliminated competition and 13 held prices for cable services above competitive levels. Id. at 29-30. The plaintiffs 14 proposed four theories of individual injury resulting from the alleged antitrust violation, 15 but the district court accepted only one of the theories as capable of classwide proof. Id. 16 at 31. The plaintiffs’ expert designed a regression model comparing actual cable prices in 17 the region with hypothetical prices that would have prevailed but for the defendants’ 18 allegedly anticompetitive activities. The model assumed the validity of all four theories of 19 alleged harm and did not isolate damages resulting from any one theory. Id. at 32, 36. In 20 contrast, there is only one theory of liability here: “Plaintiff’s liability theory is ‘an inflated 21 stock price as a result of certain misrepresentations and/or omissions’ and corrective 22 disclosures that removed such inflation.” (Doc. 168 at 16.) 23 “[U]ncertain damages calculations do not alone defeat certification,” and 24 “individualized damages issues do not alone defeat certification.” Huu Nguyen v. Nissan 25 N. Am., Inc., 932 F.3d 811, 817 (9th Cir. 2019). But plaintiffs must be able to show that 26 their damages stemmed from the defendants’ actions that created the legal liability. They 27 also must be able to show that damages are capable of measurement on a classwide basis, 28 i.e., the entire class suffered damages traceable to the same injurious actions underlying

- 27 - 1 the plaintiffs’ legal theory. Id. In Nguyen, the plaintiff’s legal theory was that the 2 defendant knowingly designed a defective clutch system and did not inform consumers of 3 the defect. Id. at 819. Under this theory, the defect existed in each of the class vehicles at 4 the time of purchase and required remedy regardless of whether class members had 5 experienced performance problems caused by the defect. Thus, the appropriate measure 6 of damages was the benefit of the bargain and could be applied classwide. However, if the 7 plaintiff had alleged that performance problems constituted the defect, class members 8 might have received varying levels of compensation based on if and when they experienced 9 performance problems. Id. 10 Defendants contend that Plaintiff has not satisfied Rule 23(b)(3)’s predominance 11 requirement to proffer a methodology for calculating classwide damages that is consistent 12 with his theory of liability. Coffman, Plaintiff’s expert, opined that “damages in this action 13 are subject to a well-settled, common methodology that can be applied to the class as a 14 whole.” (Doc. 160-3 at 36.) He stated: 15 [T]he standard and well-settled formula for assessing damages for each class member under Section 10(b) is the “out-of-pocket” method which measures 16 damages as the artificial inflation per share at the time of purchase less the 17 artificial inflation at the time of sale (or, if the share is not sold before full revelation of the fraud, the artificial inflation at the time of purchase, subject 18 to the PSLRA’s “90-day lookback” provision, a formulaic limit on damages 19 that also can be applied class-wide).

20 .... The methodology and evidence for establishing the artificial inflation per 21 share in the market price on each day during the Class Period is also common 22 to the class and can be measured class-wide. In particular, as is standard procedure in Section 10(b) cases, the most common methodology to quantify 23 artificial inflation is to perform an event study that measures price reactions 24 to disclosures that revealed the relevant truth concealed by the alleged material omissions and/or misrepresentations. This analysis, and the 25 evidence supporting it, would be common to the class. 26 (Id. at 35.) 27 28

- 28 - 1 Defendants do not dispute that the standard “out-of-pocket” method and an event 2 study are appropriate methods for determining damages in a securities fraud case. They 3 do not contend that individual questions predominate. Rather, they contend that Coffman 4 has not explained how his study would isolate the decline in Insys’s stock price that is 5 allegedly attributable to the two misstatements at issue. Smith, Defendants’ expert, opined 6 that Coffman’s anticipated “analysis will be insufficient to isolate only the price effect of 7 the allegedly corrective information from other potentially confounding news that 8 impacted the price of Insys stock but was not related to the allegations in this matter.” 9 (Doc. 165-2 at 29-30.) As an example, Smith said that on December 3, 2015, an alleged 10 corrective disclosure occurred and also Insys held an “investor day,” during which senior 11 management provided information that investors would likely find relevant to the 12 company’s stock price. Smith commented that Coffman had not addressed how he would 13 separate the effect of the “investor day” news from that of the alleged corrective disclosure 14 the same day. (Id. at 30.) Coffman responded: 15 81. Dr. Smith does not dispute that out-of-pocket damages in a class action pursuing claims under Section 10(b) of the Securities Exchange 16 Act of 1934 can be computed based upon the artificial inflation in the stock 17 price at the time of the purchase minus the artificial inflation at the time of sale (or just the inflation at the time of purchase if the share is not sold prior 18 to revelation of the fraud). 19 82. Dr. Smith also does not dispute that the event study 20 methodology is a reliable approach to isolate the price impact of new company-specific information for a given stock on a given day, and thus, to 21 identify the potential inflation per share when corrective information is released. 22 83. Dr. Smith’s criticisms of my opinion on damages as expressed 23 in the Coffman Report do not relate to whether the general out-of-pocket 24 damages methodology can be applied formulaically in this matter using common proof on a class-wide basis. Instead, his objections all relate to 25 whether certain adjustments would ultimately be necessary as a matter of loss 26 causation at the merits stage to address (1) the purported release of putatively confounding information, if any, on the alleged corrective disclosure dates; 27 and (2) how far back in time any inflation per share extends, i.e., when 28 damages, and the Class Period, started. Answering these questions requires

- 29 - a detailed loss causation analysis, which I understand is not required at this 1 stage of the litigation, and, as a result, I have not been asked to perform. 2 84. These more detailed issues regarding loss causation (i.e., 3 “disaggregating” the effect of potential confounding information, whether in whole or in part, and how far back in time the inflation is to be carried) are 4 regularly the subject of expert and merits discovery in securities class action 5 matters, based upon a full factual record. Importantly for present purposes, however, the potential loss causation issues raised by Dr. Smith—if they are 6 pertinent at all—may be determined based upon proof that is common to all 7 class members. 85. To be clear, regardless of how the finder of fact ultimately 8 weighs the evidence and determines the appropriate percentage of the 9 abnormal stock price movement to be applied as artificial inflation resulting from the release of corrective versus any confounding information, that 10 percentage can be mechanically incorporated into the standard damages 11 model already identified using a common out-of-pocket formula. . . . In other words, regardless what the fact finder decides, the methodology for 12 calculating damages will remain the same and apply equally to each class 13 member. 14 (Doc. 169-2 at 43-44; footnotes omitted.)

15 Therefore, Plaintiff has proposed a method of calculating damages that is consistent

16 with his theory of liability and can be applied classwide. At the class certification stage,

17 he is not required to do more.

18 V. SUMMARY

19 Plaintiff has demonstrated compliance with Rule 23(a) of the Federal Rules of Civil

20 Procedure, which Defendants do not dispute. Defendants also do not dispute that Plaintiff

21 has satisfied the superiority requirement of Rule 23(b)(3). Regarding Rule 23(b)(3)’s

22 requirement that questions common to class members predominate, Plaintiff has proved

23 the prerequisites for invoking the fraud-on-the-market presumption of reliance under

24 Basic, and Defendants have not rebutted it. The Affiliated Ute presumption of reliance

25 does not apply because this case does not involve primarily a failure to disclose. Finally,

26 Plaintiff has met his burden to propose a classwide method of calculating damages that is

27 consistent with his theory of liability.

- 30 - 1 IT IS THEREFORE ORDERED that Defendants’ Motion for Leave to File Sur- 2 Reply in Further Opposition to Lead Plaintiff’s Motion for Class Certification (Doc. 170) 3 is granted. The Clerk will file the sur-reply attached as Exhibit A to the motion (Doc. 170- 4 1). 5 IT IS FURTHER ORDERED that Lead Plaintiff’s Motion for Class Certification 6 (Doc. 159) is granted. 7 IT IS FURTHER ORDERED that this action is certified as a class action pursuant 8 to Federal Rules of Civil Procedure 23(a) and 23(b)(3) on behalf of a Class consisting of 9 all persons and entities who purchased or otherwise acquired Insys Therapeutics, Inc., 10 common stock during the period from March 3, 2015, through January 25, 2016, and were 11 damaged thereby. Excluded from the Class are (a) Defendants; (b) present and former 12 directors or executive officers of Insys and members of their immediate families (as defined 13 in 17 C.F.R. § 229.404, Instructions (1)(a)(iii) and (1)(b)(ii)); (c) any of the foregoing 14 individuals’ or entities’ legal representatives, heirs, successors, or assigns; and (d) any 15 entity in which any Defendant has or had a controlling interest, or which is related to or 16 affiliated with, any Defendant. 17 IT IS FURTHER ORDERED that Clark Miller is appointed as Class Representative. 18 IT IS FURTHER ORDERED that, pursuant to Federal Rule of Civil Procedure 19 23(g), Kessler Topaz Meltzer & Check, LLP, is appointed as Class Counsel and Bonnett, 20 Fairbourn, Friedman & Balint, P.C., is appointed as Liaison Counsel. 21 Dated this 20th day of September, 2019. 22 23 24 25 26 27 28

- 31 -