Xu v. FibroGen, Inc.

Opinion

PEIFA XU, Case No. 21-cv-02623-EMC

Plaintiff, ORDER GRANTING MOTIONS TO v. CONSOLIDATE; AND GRANTING RETIREMENT SYSTEMS’ MOTIONS FIBROGEN, INC., et al., FOR APPOINTMENT AS LEAD PLAINTIFF AND APPROVAL OF Defendants. LEAD COUNSEL

Docket Nos. 22, 29, 40

This case is a securities-fraud class action brought on behalf of investors who purchased stock in FibroGen, Inc., from October 2017 through April 2021. Plaintiffs assert claims under Sections 10(b) and 20(a) of the Securities Exchange Act of 1934 (the “Exchange Act”) as well as Securities and Exchange (“SEC”) Rule 10b-5. Pending before the Court are three class members’ motions for consolidation of related actions, appointment as lead plaintiff, and approval of lead counsel pursuant to the Private Securities Litigation Reform Act of 1995 (“PSLRA”). The motions were filed by Plaintiffs Vicente Sepulveda, see Docket No. 22 (“Sepulveda Mot.”); the Employees’ Retirement System of the City of Baltimore (the “Baltimore Fund”), the City of Philadelphia Board of Pensions and Retirement (the “Philadelphia Fund”), and the Plymouth County Retirement Association (the “Plymouth Fund”) (collectively, the “Retirement Systems”), see Docket No. 29 (“Ret. Sys. Mot.”); and Stefano Branca and Giuliana Mollo, see Docket No. 40 (“Branca-Mollo Mot.”). For the reasons given below, the Court GRANTS the parties’ motions to consolidate the as lead plaintiffs and approval of its selected law firm, Saxena White, as lead counsel. A. Factual Background Plaintiff Peifa Xu filed a class action complaint in this Court on April 12, 2021. See Docket No. 1 (“Compl.”). According to the Xu complaint, Defendant FibroGen “is a biopharmaceutical company that develops medicines for the treatment of anemia, fibrotic disease, and cancer.” Id. ¶ 20. “Its most advanced product is roxadustat,” an oral medication “for the treatment of anemia due to chronic kidney disease (‘CKD’).” Id. ¶ 20. In November 2019, “FibroGen issued a press release announcing ‘Positive Phase 3 Pooled Roxadustat Safety and Efficacy Results’” based on six global clinical trials. Id. ¶ 25. The press release specifically stated that roxadustat “demonstate[d] a cardiovascular safety profile comparable with placebo in patients not on dialysis, and comparable or in some cases better than that of epoetin alfa in patients on dialysis.”1 Id. The following month, in December 2019, “the Company filed its New Drug Application (‘NDA’) with the U.S. Food and Drug Administration (‘FDA’) for the approval of roxadustat.” Id. ¶ 20, 26. In its press release announcing the NDA submission, FibroGen again touted “positive results from a global Phase 3 program encompassing 15 trials that enrolled more than 10,000 patients, worldwide.” Id. ¶ 26. From February to December 2020, the company made additional public statements suggesting that the FDA review process was proceeding smoothly. See id. ¶¶ 27-28. In April 2021, however, FibroGen issued a press release that “provided clarification of certain prior disclosures of U.S. primary cardiovascular safety analyses from the roxadustat Phase 3 program.” Id. ¶ 30. According to the statement, senior management became aware, while “preparing for [an] upcoming FDA Advisory Committee meeting,” that the earlier “cardiovascular safety analyses included post-hoc changes to . . . stratification factors.” Id. When these changes were removed, “the pre-specified stratification factors result[ed] in higher hazard ratios” such that FibroGen could no longer represent that roxadustat is safer than epoetin alfa in treating CKD anemia. See id. ¶¶ 30, 32. The following two days, “the Company’s share price fell $14.90, or 43%.” Id. ¶ 31. The Xu complaint alleges that FibroGen’s public statements prior to the April 2021 disclosure “were materially false and/or misleading, and failed to disclose material adverse facts about the Company’s business, operations, and prospects.” Id. ¶ 29; see also id. (specifying the ways in which FibroGen’s statements misled investors and/or lacked a reasonable basis). The complaint asserts one claim under Section 10(b) of the Exchange Act and SEC Rule 10b-5 against all Defendants and another under Section 20(a) of the Exchange Act against individual Defendants Enrique Contero, James Schoeneck, and K. Peony Yu, who were officers of the company at relevant times. See id. ¶¶ 8-12, 38-51. B. Procedural Background After the Xu complaint was filed in this Court, similar actions were brought by purchasers of FibroGen securities elsewhere in this district. See Gutman v. FibroGen, Inc., No. 3:21-cv- 02725-YGR; Grazioli v. FibroGen, No. 3:21-cv-03212-CRB; IBEW Local 353 Pension Plan v. FibroGen, Inc., No. 3:21-cv-03396-EJD; Leonard v. FibroGen, Inc., No. 3:21-cv-03370-EMC. The Class Period asserted in the Leonard action is the longest and runs from October 18, 2017, through April 6, 2021.2 Beginning on June 11, 2021, five class members filed motions for consolidation of related actions, appointment as lead plaintiff, and approval of lead counsel. They include Plaintiffs Brett Richard (Docket No. 18), Sepulveda (Docket No. 22), the Retirement Systems (Docket No. 29), Thomas Leonard (Docket No. 37), and Branca-Mollo (Docket No. 40). On June 14, 2021, Plaintiff Richard withdrew his earlier motion. See Docket No. 48. On June 25, 2021, Plaintiff Leonard filed a notice of non-opposition to the competing motions. See Docket No. 49. Also on June 25, 2021, Sepulveda, the Retirement Systems, and Branca-Mollo each filed oppositions to

2 The movants agree that “[f]or the purpose of determining lead plaintiff, . . . use of the longer, most inclusive class period . . . is proper, as it encompasses more potential class members.” See Ali v. Intel Corp., 2018 WL 2412111, at *2 n.6 (N.D. Cal. May 29, 2018) (quoting In re Doral one another’s competing motions. See Docket Nos. 50 (“Sepulveda Opp’n”), 52 (“Ret. Sys. Opp’n”), and 51 (“Branca-Mollo Opp’n”), respectively. On July 2, 2021, Sepulveda, the Retirement Systems, and Branca-Mollo filed reply briefs in support of their motions. See Docket Nos. 54 (“Sepulveda Reply”), 57 (“Ret. Sys. Reply”), and 56 (“Branca-Mollo Reply”), respectively. The Court held a hearing on the competing motions on August 19, 2021. See Docket No. 74. C. Consolidation of Related Actions Where more than one action “on behalf of a class asserting substantially the same claim or claims arising under” the PSLRA “has been filed, and any party has moved to consolidate those actions,” the motion to consolidate must be decided prior to appointment of a lead plaintiff. 15 U.S.C. § 78u–4(a)(3)(B)(ii). Under Federal Rule of Civil Procedure 42(a), a court may consolidate two or more actions that “involve a common question of law or fact.” Fed. R. Civ. P. 42(a). “The purpose of consolidation is to avoid the unnecessary costs or delays that would ensue from proceeding separately with claims or issues sharing common aspects of law or fact.” Weisz v. Calpine Corp., 2002 WL 32818827, at *2 (N.D. Cal. Aug. 19, 2002) (internal citation omitted). “[S]o long as any confusion or prejudice does not outweigh efficiency concerns, consolidation will generally be appropriate.” Primavera Familienstiftung v. Askin, 173 F.R.D. 115, 129 (S.D.N.Y. 1997). Securities actions are often consolidated, especially where they are “based on the same public statements and reports.” See id. (internal quotation omitted). All three movants agree that the five related actions should be consolidated. See Sepulveda Mot. at 5, Ret. Sys. Mot. at 7-8, Branca-Mollo Mot. at 3. They argue that these actions “present virtually identical factual and legal issues,” as each “alleges violations of Sections 10(b) and 20(a) of the Exchange Act, names common Defendants, and stems from the same underlying operative facts and circumstances.” See Sepulveda Mot. at 5. Additionally, “[t]he misstatement theories of the five cases are nearly identical,” with complaint alleging that “Defendants made materially false and/or misleading statements and/or failed to disclose material adverse facts about the Phase 3 roxadustat trial”; specifically, “[t]he complaints all allege Defendants’ misstatements ¶ 29, Gutman Compl. ¶ 21, Grazioli Compl. ¶ 22, IBEW Compl. ¶ 33, and Leonard Compl. ¶ 67). Neither FibroGen nor the individual Defendants oppose the motions to consolidate. Because consolidation would “expedite these proceedings, reduce duplicative efforts, and save all parties time, energy, and expense,” the Court GRANTS the plaintiffs’ motion for consolidation of the related actions. See Sepulveda Mot. at 5. D. Appointment as Lead Plaintiff The PSLRA provides that plaintiffs bringing a securities class action must publish, “[n]ot later than 20 days after the date on which the complaint is filed . . . in a widely circulated national business-oriented publication or wire service, a notice advising members of the purported plaintiff class” of “the pendency of the action, the claims asserted therein, and the purported class period,” and informing prospective class members “that, not later than 60 days after the date on which the notice is published, any member of the purported class may move the court to serve as lead plaintiff of the purported class.” 15 U.S.C. § 78u–4(a)(3)(A)(i). Not later than 90 days after the date on which a notice is published . . . the court shall consider any motion made by a purported class member in response to the notice” to serve as lead plaintiff. 15 U.S.C. § 78u–4(a)(3)(B)(i). Courts considering motions for appointment as lead plaintiff “shall adopt a presumption that the most adequate plaintiff . . . is the person or group of persons that

(aa) has either filed the complaint or made a motion in response to a notice under subparagraph (A)(I); (bb) in the determination of the court, has the largest financial interest in the relief sought by the class; and

(cc) otherwise satisfies the requirements of Rule 23 of the Federal Rules of Civil Procedure. 15 U.S.C. § 78u–4(a)(3)(B)(iii)(I). In accordance with these provisions, the Ninth Circuit has instructed district courts deciding motions for appointment as lead plaintiff to follow a three-step process. See In re Cavanaugh, 306 F.3d 726, 729-30 (9th Cir. 2002). The first step consists in determining whether been satisfied. See id. at 729. At step two,

the district court must compare the financial stakes of the various plaintiffs and determine which one has the most to gain from the lawsuit. It must then focus its attention on that plaintiff and determine, based on the information he has provided in his pleadings and declarations, whether he satisfies the requirements of Rule 23(a), in particular those of “typicality” and “adequacy.” If the plaintiff with the largest financial stake in the controversy provides information that satisfies these requirements, he becomes the presumptively most adequate plaintiff. If the plaintiff with the greatest financial stake does not satisfy the Rule 23(a) criteria, the court must repeat the inquiry, this time considering the plaintiff with the next-largest financial stake, until it finds a plaintiff who is both willing to serve and satisfies the requirements of Rule 23. Id. at 730 (emphasis in original) (footnotes omitted). “The third step of the process is to give other plaintiffs an opportunity to rebut the presumptive lead plaintiff's showing that it satisfies Rule 23’s typicality and adequacy requirements.” Id. (citing 15 U.S.C. § 78u-4(a)(3)(B)(iii)(II)).

If, as a result of this process, the district court determines that the presumptive lead plaintiff does not meet the typicality or adequacy requirement, it then must proceed to determine whether the plaintiff with the next lower stake in the litigation has made a prima facie showing of typicality and adequacy. If so, it must declare that plaintiff the presumptive lead plaintiff and repeat step three of the process by giving other plaintiffs an opportunity to rebut that showing. This process must be repeated sequentially until all challenges have been exhausted. See id. Id. at 731 (emphasis in original) (footnote omitted).3 1. Notice Requirement As stated above, the Xu complaint was filed in this Court on April 12, 2021. See Docket No. 1. The same day, “a notice (‘Notice’) was published on Newsfile that notified members of the proposed class that a class action lawsuit had been initiated against FibroGen and certain of its

3 Commenting on the overlap between steps two and three, the Cavanaugh court acknowledged that “it may seem incongruous to allow other plaintiffs to present evidence casting doubt on the determination just made by the district court that the presumptive lead plaintiff satisfies Rule 23’s adequacy and typicality requirements.” Cavanaugh, 306 F.3d at 730. The Ninth Circuit clarified, however, that the district court’s initial determination at step two must be based “on the presumptive lead plaintiff’s complaint and sworn certification; there is no adversary process to test the substance of those claims.” Id. “At the third stage,” in contrast, “the process turns adversarial executives, and that members of the asserted class had a right to file a motion seeking appointment as Lead Plaintiff within 60 days of the publication of the Notice, i.e., on or before June 11, 2021.” Ret. Sys. Mot. at 9; see also Docket Nos. 23-1, 30-1, 41-1 (comprising the notice of pendency in Newsfile). Sepulveda, the Retirement Systems, and Branca-Mollo all filed their motions for appointment as lead plaintiff on June 11, 2021, or sixty days after publication of the Notice. Because the Notice was published within twenty days of the Xu complaint’s filing and the potential plaintiffs’ motions were in turn filed within sixty days of the Notice’s publication, the movants have satisfied the procedural requirements of the PSLRA. See Sepulveda Mot. at 6, Ret. Sys. Mot. at 9, Branca-Mollo Mot. at 4. 2. Movants’ Financial Interests “The Ninth Circuit has not provided clear guidance on what metric district courts should use in determining which potential plaintiff has the largest financial interest in a case, noting only that ‘the court may select accounting methods that are both rational and consistently applied.’” Mulligan v. Impax Labs., Inc., 2013 WL 3354420, at *4 (N.D. Cal. July 2, 2013) (quoting Cavanaugh, 306 F.3d at 730 n.4). In Melucci v. Corcept Therapeutics Inc., 2019 WL 4933611 (N.D. Cal. Oct. 7, 2019), Judge Koh observed that “[t]o evaluate approximate economic losses, courts frequently weigh four factors that were first introduced in Lax v. First Merchants Acceptance Corp., 1997 WL 461036, at *5 (N.D. Ill. Aug. 11, 1997) and In re Olsten Corp. Securities Litigation, 3 F. Supp. 2d 286, 295 (E.D.N.Y. 1998).” See 2019 WL 4933611 at *3. “Under the Lax-Olsten test, courts consider: (1) the number of shares purchased during the class period; (2) the number of net shares purchased during the class period; (3) the total net funds expended during the class period; and (4) the approximate losses suffered during the class period.” Id. (internal quotation omitted). While “[c]ourts applying this test consider the factors together as a whole to approximate each moving plaintiff’s losses,” id. (citing Vancouver Alumni Asset Holdings, Inc. v. Daimler AG, 2016 WL 10646304, at *2-3 (C.D. Cal. July 20, 2016)), they “generally place the greatest emphasis on the last of these factors,” i.e., the approximate losses suffered, City of Royal Oak Ret. Sys. v. Juniper Networks, Inc., 2012 WL 78780, at *4 (N.D. Cal. In evaluating plaintiffs’ financial interests in the litigation, moreover, courts in this district and elsewhere have often distinguished between plaintiffs’ “actual economic losses suffered” and their “potential recovery” via federal securities laws. See Perlmutter v. Intuitive Surgical, Inc., 2011 WL 566814, at *3 (N.D. Cal. Feb. 15, 2011). As Judge Koh observed in Perlmutter, “the distinction between these two categories is an artificial one” insofar as “both categories should yield the same result,” at least in principle. Id. at *4. The Supreme Court’s ruling in Dura Pharmaceuticals, Inc. v. Broudo, 544 U.S. 336 (2005), however, “made it clear that a purchaser of stock at fraudulently inflated prices may suffer economic losses that are not caused by a defendant’s misrepresentations,” and are therefore not recoverable. See 2011 WL 566814 at *4; see also Dura, 544 U.S. at 342-43 (stating that an investor’s losses on sales of stock whose purchase price had been fraudulently inflated “may reflect, not the earlier misrepresentation, but changed economic circumstances . . . or other events, which taken separately or together account for some or all of th[e] lower price”; consequently, where an investor sells his “shares quickly before the relevant truth begins to leak out, the misrepresentation will not have led to any loss”). In response to Dura, many district courts—including this one—have chosen “not to consider losses resulting from stock trades that occurred prior to any disclosure of the defendant’s fraud” when evaluating potential plaintiffs’ financial interests in the litigation.4 See Perlmutter, 2011 WL 566814, at *4. Here, the Court finds that the Retirement Systems have the greatest financial interest in the litigation, whether measured as actual losses suffered or potential recovery under Dura.5 In their 4 In Perlmutter, Judge Koh concluded that “Ninth Circuit authority favors considering loss causation on a motion for appointment as lead plaintiff,” and that courts should therefore focus only on “losses occurring after a defendant makes corrective disclosures.” Id. at *5-6 (citing, e.g., Cavanaugh, 306 F.3d at 730, which stated that district courts should determine “which [potential plaintiff] has the most to gain from the lawsuit”) (emphasis added). In Mulligan, this Court similarly suggested that courts should calculate losses incurred by plaintiffs in light of Dura, such that “losses stem[ming] from stocks sold before the alleged fraud was revealed” are not considered. See 2013 WL 3354420, at *5-6; see also Isaacs v. Musk, 2018 WL 6182753, at *2 (N.D. Cal. Nov. 27, 2018) (declining to appoint as lead plaintiff an investor whose claimed losses were largely attributable to activity occurring “prior to the fraud” and so potentially not “causally related to the fraud”). motions and accompanying loss charts, the movants calculate their respective losses in the following manner: Shares Net Shares Net Funds Approximate Purchased Purchased Expended Losses6 Sepulveda7 n/a n/a $119,030 $946,647 Common 209,548 (11,900) ($460,668) $300,500 Stock Stock Options 9,207 (247) $579,698 $646,148 Retirement 71,477 49,318 $2,340,770 $1,330,943 Systems8 Baltimore n/a n/a n/a $505,254 Fund Philadelphia n/a n/a n/a $475,238 Fund Plymouth n/a n/a n/a $350,451 Fund Branca-Mollo9 n/a 21,549 n/a $459,057 thus, causally connected to the fraud.” Ret. Sys. Opp’n at 9 n.5 (emphasis in original). The competing movants do not challenge this calculation or the broader point that the vast majority of the Retirement Systems’ losses is presumably cognizable under Dura.

6 The approximate losses provided here are based on a “last-in-first-out” (“LIFO”) accounting methodology, rather than a “first-in-first-out” (“FIFO”) methodology, as the movants agree that a LIFO calculation is most appropriate here. See In re Cloudera, Inc. Sec. Litig., 2019 WL 6842021, at *4 (N.D. Cal. Dec. 16, 2019) (stating that application of a LIFO method “comports with the weight of authority and avoids artificially inflating losses with market activity outside of the class period”).

7 See Sepulveda Mot. at 7, Docket No. 23-3. Sepulveda’s figures combine his interests in FibroGen common stock and common stock options. See Docket No. 23-3 at 2. As Sepulveda’s loss chart indicates, he was both a net seller and a net gainer in his transactions in FibroGen common stock.

8 See Ret. Sys. Mot. at 9-10, Docket No. 30-3. The Baltimore Fund’s approximate losses of $505,254 includes $203,190 that is attributable to an assignment of claims in this matter from a sister fund, the Fire & Police Employees’ Retirement System of the City of Baltimore. See Docket No. 30-3 at 2. This assignment is discussed further below. Based the movants’ own representations, the Retirement Systems—when aggregating the losses of their constituent members—have suffered total losses of more than $1.3 million, while Sepulveda has suffered losses of approximately $947,000 and Branca-Mollo have suffered losses of around $459,000. The Retirement Systems therefore have “the largest financial interest in the relief sought by the class,” 15 U.S.C. § 78u-4(a)(3)(B)(iii)(I)(bb), and are presumptively the most adequate plaintiff when considered on that basis. In opposing the Retirement Systems’ motion, Sepulveda implies that the Baltimore F&P’s losses should not be considered in calculating the Retirement Systems’ interest in the litigation.10 Specifically, he argues that one of the Retirement Systems’ members, the Baltimore Fund, submitted a flawed PSLRA certification because that document refers to losses of both the Baltimore Fund proper and another entity, the Fire & Police Employees’ Retirement System of the City of Baltimore (“Baltimore F&P”). Sepulveda Opp’n at 7-8, Docket No. 30-2 at 2. The certification states that David Randall, the Executive Director of the Baltimore Fund, is authorized to engage in this litigation on behalf of the Baltimore F&P “pursuant to an assignment of claims from Baltimore F&P to Baltimore Employees.”11 Docket No. 30-2 at 2. Sepulveda contends that

Mollo identify their retained shares, a similar method for calculating losses in securities cases. On the similarities and differences between the two approaches, see Mulligan, 2013 WL 3354420, at *6-8. The Retirement Systems have calculated the shares purchased and net funds expended for Branca-Mollo and assert that they purchased 21,549 total shares and expended $548,973 in net funds. See Ret. Sys. Opp’n at 5.

10 Sepulveda’s arguments about the Baltimore F&P are presented as attacks on the Retirement Systems’ adequacy and typicality, as Sepulveda contends that the Retirement Systems are subject to unique defenses. Sepulveda Opp’n at 7-8. The Court considers it appropriate to consider these arguments as part of the financial-interest calculation, as the Baltimore F&P specific losses seem unlikely to otherwise play a significant role in the litigation. See, e.g., Faris v. Longtop Fin. Techs. Ltd., 2011 WL 4597553, at *5-7 (S.D.N.Y. Oct. 4, 2011) (finding that a lead plaintiff received a valid assignment from a third party and rejecting a competing movant’s argument that the assignment would subject the lead plaintiff to a unique defense).

11 The Retirement Systems also filed a declaration from the Executive Director of the Baltimore F&P stating that the organization “has resolved to assign to [the] Baltimore [Fund] . . . all rights, title, and interest in any and all claims, demands, and causes of action of any kind whatsoever arising from violations of the U.S. federal securities laws . . . as may be asserted against FibroGen and related parties in connection with Baltimore F&P’s purchase of FibroGen securities.” Docket No. 30-5 at 1-2; see also Ret. Sys. Reply at 10 (reiterating that the Baltimore Fund “received by Randall “does not indicate that he had any role with Baltimore F&P during the Class Period” and that the certification therefore “raises serious questions concerning the enforceability of the purported assignment of Baltimore F&P’s claims.”12 Sepulveda Opp’n at 7-8. The Retirement Systems respond that the two Baltimore entities are “sister fund[s]” that share “a common investment manager,” have “offices in the same building,” and have been managed by some of the same individuals. Ret. Sys. Reply at 9 & n.8. They further point out that “[a]ssignments of claims have long been held valid in Maryland, as elsewhere,” id. (citing Med. Mut. Liab. Ins. Soc. of Md. v. Evans, 622 A.2d 103, 116 (1993)), and that “courts in the Ninth Circuit and around the country frequently appoint lead plaintiffs who received all or a portion of their claims via assignment,” id.; see, e.g., Rieckborn v. Velti PLC, 2013 WL 6354597, at *3 (N.D. Cal. Dec. 3, 2013) (appointing an assignee as lead plaintiff and collecting additional cases “holding that . . . formal assignments of legal claims could confer . . . standing” on third parties to pursue securities claims). The Court finds that the Baltimore F&P’s assignment of its claims in this matter to the Baltimore Fund is facially valid and does not raise concerns significant enough to warrant subtracting the Baltimore F&P’s losses from those of the Retirement Systems as a whole for purposes of appointing a lead plaintiff. There is no substantial evidence contesting the bona fides of the assignment. In any event, even if the Court were to disaggregate the Baltimore F&P’s losses of $203,190 from the Retirement Systems’ losses, the Retirement Systems would still have incurred losses of well over $1 million, and thus remain the movant with the greatest financial interest in the litigation.13 12 The Retirement Systems’ certification also states that the Baltimore Fund has powers-of- attorney with respect to Baltimore F&P in this litigation, Docket No. 30-5 at 2, and Sepulveda questions whether this provision, either separately or in tandem with the assignment provision, complies with the notarization and attestation requirements of Maryland law. Sepulveda Opp’n at 7-8. In response, the Retirement Systems submitted a new power-of-attorney agreement in advance of the motion hearing that appears to resolve Sepulveda’s concerns. See Docket No. 73- 1. The Court therefore considers this issue moot.

13 While the Court need not address the financial interest of the other movants unless if finds that the Retirement Systems fail to satisfy the requirements of Rule 23, it notes here that both the Retirement Systems and Branca-Mollo raise serious questions about whether Sepulveda’s asserted losses of approximately $946,647 are recoverable under Dura. They argue that “Sepulveda should 3. Typicality and Adequacy At step two of the PSLRA’s three-step process, courts must also ask whether the movant with the largest financial stake has made a prima facie showing, through its “pleadings and declarations,” that it satisfies the typicality and adequacy requirements of Rule 23(a). See 15 U.S.C. § 78u–4(a)(3)(B)(iii)(II)(aa)-(bb); Cavanaugh, 306 F.3d at 732. “The test for typicality asks whether other members have the same or similar injury, whether the action is based on conduct which is not unique to the named plaintiffs, and whether other class members have been injured by the same course of conduct.” Melucci, 2019 WL 4933611, at *4 (internal quotation omitted). “The test for adequacy asks whether “the movant and its counsel have any conflicts of interest with other class members” and whether “the movant and its counsel [will] prosecute the action vigorously on behalf of the class.” In re Mersho, --- F.4th ---, 2021 WL 3121385, at *5 (9th Cir. July 23, 2021) (internal quotation omitted). As discussed above, the process “turns adversarial” at step three, and competing movants may “rebut the presumptive lead plaintiff's showing that it satisfies Rule 23’s typicality and adequacy requirements.” Id. at 730. This is done by “prov[ing]” that the presumptive lead plaintiff “is subject to unique defenses that render [it] incapable of adequately representing the class” or that it “will not fairly and adequately protect the interests of the class.” See 15 U.S.C. § 78u–4(a)(3)(B)(iii)(II)(aa)-(bb). In evaluating a potential lead plaintiff’s adequacy, courts throughout this district have often expressed skepticism toward artificial, or lawyer-engineered, groups. See, e.g., In re Stitch Fix, Inc. Sec. Litig., 393 F. Supp. 3d 833, 835 (N.D. Cal. 2019) (“Although there is no controlling amounts, they assert, “the vast majority of [his] loss” would be erased. Id. Branca and Mollo thus calculate Sepulveda’s potential Dura recovery to be approximately $143,000, see Branca-Mollo Reply at 3, while the Retirement Systems posit that his cognizable losses amount to only $98,405. See Ret. Sys. Opp’n at 7-9. Sepulveda’s essential counterargument is that the Court should “decline to apply a loss causation methodology premised on Dura at the lead plaintiff appointment stage,” as “doing so requires the court to prematurely delve into a complicated merits analysis where expert opinion is needed.” See Sepulveda Reply at 7. But, as explained above, this Court and many others have concluded that they “would be abdicating [their] responsibility under the PSLRA if [they] were to ignore the issue of loss causation at the lead plaintiff appointment stage.” See Africa v. Jianpu Tech. Inc., 2021 WL 1999467, at *2 (S.D.N.Y. May 19, 2021). The Court thus reaffirms, especially given the likelihood that the bulk of Sepulveda’s claimed losses would not be recoverable under a Dura analysis, that the Retirement Systems’ financial stake in the authority on point, the clear consensus in our district is that a group of investors who had no pre- existing relationship with one another, and whose relationship and group status were forged only by a lawyer, is not appropriate to be lead plaintiff based on their aggregated losses.”). This Court recently elaborated on the underlying concerns with artificial groups in Haideri v. Jumei International Holding Ltd., 2020 WL 5291872 (N.D. Cal. Sept. 4, 2020). It explained:

[C]ourts have often looked with skepticism at “artificial groups,” i.e., groups made up of persons or entities that did not have a pre- litigation relationship. . . .

Underlying this skepticism is that such groups are often formed by lawyers for purposes of obtaining appointment as lead counsel. Appointing a lawyer-engineered group would undercut[] the primary purpose of the PSLRA: to eliminate lawyer-driven litigation. . . . Moreover, allowing for appointment of lawyer-engineered groups may lead to unintended results such as generating a flurry of otherwise pointless activity that adds nothing to the prompt and fair resolution of disputes and creating powerful incentives for lawyers competing to represent the class to solicit clients. Id. at *3-4 (internal quotations omitted) (emphasis in original); see also In re Network Assocs., Inc., Sec. Litig., 76 F. Supp. 2d 1017, 1019-27 (N.D. Cal. 1999) (Alsup, J.) (providing a detailed discussion of the drawbacks of “artificial groups” in the PSLRA context and concluding that courts should “consider a group candidate [as lead plaintiff] only if it meets . . . strict criteria”). Given these concerns, this Court in Jumei considered the following factors in deciding whether to allow the artificial group to serve as lead plaintiff: “(1) the existence of a pre-litigation relationship between group members; (2) involvement of the group members in the litigation thus far; (3) plans for cooperation; (4) the sophistication of its members; and (5) whether the members chose outside counsel, and not vice versa.” Id. at *4 (quoting Brady v. Top Shops Inc., 324 F. Supp. 3d 335, 345 (E.D.N.Y. 2018)). The Court ultimately found that the group was not adequate to serve as lead plaintiff, as its members provided only conclusory “information about themselves (e.g., investment experience), their relationships with one another, how they came to be formed as a group, and (perhaps most significantly) how they would cooperate and manage the litigation given their status as a group.” Id. at *5. Mersho. There, the district court found that such a group had the largest financial interest of all potential plaintiffs and that it “made a prima facie showing of adequacy and typicality.” 2021 WL 3121385, at *2. But the district court also averred that courts “uniformly refuse to appoint groups of unrelated investors who are brought together for the sole purpose of aggregating their claims in an effort to become the presumptive lead plaintiff.” Id. (internal quotations omitted). As a result, the district court declined to appoint the group lead plaintiff on its own initiative and despite its finding of prima facie adequacy. Id. at *3, *5-6. On appeal, the Ninth Circuit held that the district court erred because it “effectively left the burden on [the group] to prove adequacy at step three even though the burden should have shifted to the competing movants to show inadequacy.” Id. at *6. Nevertheless, the Ninth Circuit affirmed that district courts have “latitude as to what information [they] will consider in determining typicality and adequacy” and that they are not “precluded from considering pre-litigation relationships or cohesion” as part of their adequacy analysis. Id. at *5-6 (quoting Cavanaugh, 306 F.3d at 732). “District courts,” the Ninth Circuit went on, “often consider a pre-litigation relationship along with other factors such as the size of the group, how the members found their counsel, and the prosecution procedures set out in their filings.”14 Id. at *6. Mersho therefore indicates that district courts may consider the aforementioned problems associated with artificial groups in their adequacy analysis, both at step two (which requires a prima facie showing of adequacy) and at step three (which permits competing movants to rebut the first movant’s prima facie showing).15 14 In listing these factors, the court approvingly cited In re Cloudera, Inc. Sec. Litig., 2019 WL 6842021 (N.D. Cal. Dec. 16, 2019), where Judge Koh discussed a number of cases from this district, including Isaacs, in holding that groups of unrelated individuals and/or entities often make unsuitable lead plaintiffs. See id. at *7 (citing Isaacs, 2018 WL 6182753, at *3) (internal quotation omitted).

15 Some district courts have effectively disqualified artificial groups from appointment as lead plaintiffs by refusing to aggregate the group members’ losses at the previous step in the analysis, i.e., when determining the movants’ financial interest in the litigation. See, e.g., In re Stitch Fix, 393 F. Supp. 3d at 836 (“To permit aggregation and lead plaintiff status for [an artificial] group undercuts the goal of having the plaintiffs and not the lawyers can the shots in securities class actions. . . . Consequently, the Court declines to aggregate the Stitch Fix Investor Group members’ losses or to consider their individual losses.”); Koffsmon v. Green Dot Corp., 2021 WL 3473975, at *2-3 (C.D. Cal. Aug. 6, 2021) (“[T]o allow [an artificial group] to aggregate its members’ losses As the Retirement Systems have the largest financial interest in the instant litigation, the Court next considers whether they have made a prima facie showing of typicality and adequacy and, if so, whether the competing movants have rebutted that prima facie showing. As to typicality, the Retirement Systems’ own “pleadings and declarations” offer no reason to suspect that their losses are atypical of those of other class members.16 See Cavanaugh, 306 F.3d at 730. Neither Sepulveda nor Branca-Mollo meaningfully challenge this conclusion. Instead, the key issue they raise is whether the Retirement Systems constitute an improper artificial group and are therefore inadequate lead plaintiffs for purposes of Rule 23 and the PSLRA. The Court finds that the Retirement Systems have made a prima facie showing of adequacy. In terms of the factors it considered when evaluating group adequacy in Jumei, the Court emphasizes that the Retirement Systems have established (1) “a pre-litigation relationship between group members,” (2) “the sophistication of [the group’s] members,” and (3) that “the members chose outside counsel, and not vice versa.” See Jumei, 2020 WL 5291872, at *4. First, in a joint declaration submitted alongside their lead-plaintiff motion, the Retirement Systems confirm that they have prior familiarity “with one another [through] their membership in the National Conference for Public Employee Retirement Systems” as well as (in the case of the Baltimore and Philadelphia Funds) the “Mid-Atlantic Plan Sponsors, a non-profit, eleven-state organization dedicated to trustee education.” Docket No. 30-4 (“Joint Decl.”) ¶ 5. Second, the Retirement Systems credibly aver that they are all “sophisticated institutional investors” who “manage[] hundreds of millions of dollars in assets and serve[] as fiduciaries to thousands of

it was improper. See 2019 WL 6842021, at *5 (rejecting the argument “that the Court should disaggregate the potential recovery” of an artificial group since the PSLRA instructs courts to calculate the financial interest for each movant,” and instead construing the disaggregation argument “to be that the [group] has not established its adequacy to be appointed as a lead plaintiff group”). Given that the Ninth Circuit discussed many of the considerations relevant to artificial groups as part of its adequacy analysis and specifically cited Cloudera with approval, this Court also considers artificial-group issues as part of its adequacy analysis.

16 See Ret. Sys. Mot. at 11: “Like all other Class members, the Retirement Systems: (1) purchased FibroGen securities during the Class Period; (2) at prices allegedly artificially inflated by Defendants’ materially false and misleading statements and/or omissions; and (3) suffered beneficiaries,” and have extensive “prior experience . . . in selecting, hiring, and overseeing lawyers in complex litigation.” Id. Specifically, the Philadelphia and Plymouth Funds “each have significant experience serving as co-lead plaintiff[s] alongside other sophisticated investors.” Id. ¶¶ 3-4, 14; see, e.g., In re MGM Mirage Sec. Litig., No. 2:09-cv-01558 (D. Ariz.); City of Westland Police and Fire Ret. Sys. v. Sonic Solutions, No. 07-cv-05111 (N.D. Cal.). Third, the Retirement Systems state that, rather than being solicited or otherwise sought out by counsel, they first “expressed an interest to their counsel in collaborating with other sophisticated institutional investors” in this litigation and then later “independently determined to seek joint appointment as Lead Plaintiff.” Joint Decl. ¶ 8. This decision was based on the Retirement Systems’ “mutual belief that [their] partnership would allow for the sharing of experiences and resources and would add substantial value to the prosecution of the FibroGen litigation and benefit the Class.” Id. ¶ 9. Consequently, this is not a case where counsel has engineered the selection of investors and aggregated them to serve as collective plaintiffs. The competing movants argue that the Retirement Systems are inadequate because (1) the members of the Retirement Systems group “have no pre-litigation relationship,” (2) they “have exhibited minimal cooperation thus far,” and (3) they “have no coherent plan for managing the litigation.” See Sepulveda Opp’n at 5-7 (capitalization omitted). Regarding the second point, Sepulveda notes that “the total contact the group has had thus far is a conference call—presumably organized by counsel—occurring on June 10, 2021, one day before the Lead Plaintiff application filing deadline.” Id .at 6. And regarding the third point, Sepulveda contends that “the Retirement Systems do not describe any protocols they have put in place to deal with potential conflicts” beyond stating that they will resolve any disagreement by majority vote. Id. at 6-7. The competing movants argue that these deficiencies are analogous to those that this Court identified in Isaacs v. Musk, 2018 WL 6182753 (N.D. Cal. Nov. 27, 2018). There, the Court refused to appoint as lead plaintiff a group of five individuals and entities. The Court emphasized that, based on the group’s joint declaration, the group’s members (1) were “unrelated and introduced to one another by their lawyers,” (2) “participated in only one joint call prior to filing the motion for proposal that any disagreements would be resolved by majority vote. Id. at *3. The Court thus found the group artificial and lawyer-engineered, and thus unsuitable for appointment as lead plaintiff. The instant case, however, is distinguishable from Isaacs. First, as explained above, the Retirement Systems did have a “pre-litigation relationship with one another”; as Sepulveda himself acknowledges, they are all located in the northeastern United States and share a “common membership in [multiple] professional organizations.” See Sepulveda Opp’n at 5-6. Second, in addition to participating in a conference call with counsel before filing their lead-plaintiff motion, the Retirement Systems have continued to “closely track[] the progress” of this case and have “communicat[ed] both with each other and with counsel concerning specific protocols for the litigation in light of the [cost-saving] guidelines adopted by this Court” in securities class actions. Docket No. 58-1 ¶ 5; see also id. ¶¶ 5-7 (confirming that the Retirement Systems will adopt such guidelines if they are appointed lead plaintiff). These concrete measures show that the Retirement Systems have had more than de minimis contact with one another thus far and suggest that they will continue to collaborate effectively moving forward. Finally, while the Retirement Systems’ assurances that they “are committed to making all efforts . . . to reach consensus with respect to litigation decisions” and that any disagreements will be resolved “by a majority vote” are admittedly somewhat vague, the benefits that the group would bring to the class amply outweigh any deficiencies on this front. As the Court noted above, the Retirement Systems are sophisticated institutional investors with substantial experience litigating large-scale securities class actions—important attributes given “the PSLRA’s clear preference for institutional investors” to serve as lead plaintiffs. See Ret. Sys. Reply at 7; see also Jumei, 2020 WL 5291872, at *5 (“[T]he adequacy of Altimeo as a lead plaintiff is supported by the fact that it is an institutional investor.”); Knisley v. Network Assocs., Inc., 77 F. Supp. 2d 1111, 1116 (N.D. Cal. 1999) (“In enacting the PSLRA, Congress unequivocally expressed its preference for securities fraud litigation to be directed by large institutional investors.”) (internal quotation omitted). Additionally, the Retirement Systems have bound “to achieve the best possible recovery for the Class from all culpable parties.” See Joint Decl. ¶¶ 2-6. With respect to the instant case, these fiduciary responsibilities included “conduct[ing] a competitive process that vetted and evaluated potential outside counsel for securities litigation matters on the basis of, inter alia, their qualifications, experience, and fee proposals.” See Ret. Sys. Reply at 8-9. In sum, the Retirement Systems’ sophistication, litigation experience, fiduciary responsibilities to their beneficiaries, and rigorous vetting process for counsel strongly imply that they “are capable of overseeing the litigation and their proposed lead counsel in an independent manner.” Joint Decl. ¶ 14. And given that “the primary purpose of the PSLRA” is “to eliminate lawyer-driven litigation,” Jumei, 2020 WL 5291872, at *4 (internal quotation omitted) (emphasis in original), the Retirement Systems’ various attributes are sufficient to establish that they are adequate lead plaintiffs for purposes of Rule 23, as they can be expected to “prosecute the action vigorously on behalf of the class.”17 See Mersho, 2021 WL 3121385, at *5. In light of the foregoing, the Court GRANTS the Retirement Systems’ motion for appointment as lead plaintiffs and DENIES Sepulveda’s and Branca-Mollo’s corresponding motions. E. Approval of Lead Counsel Once the court decides the motion to appoint a lead plaintiff it must rule on that plaintiff’s motion for approval of class counsel. While the PSLRA requires the court to affirmatively approve the plaintiff's choice of counsel, it also “clearly leaves the choice of class counsel in the

17 The cases cited by the competing movants for the proposition that “courts consistently reject artificial institutional investor groups” as lead plaintiff, see Sepulveda Opp’n at 3, are distinguishable. See, e.g., In re Petrobras Sec. Litig., 104 F. Supp. 3d 618, 622-23 (S.D.N.Y. 2015) (denying a lead-plaintiff motion by Ohio, Idaho, and Hawaii state retirement funds where the group members had engaged in no “advance planning regarding how the litigation would be managed” and had “no agreement in place regarding the resolution of disputes amongst themselves . . . or the resolution of discrepancies in their three separate retainer agreements” with counsel”); Koffsmon, 2021 WL 3473975, at * (C.D. Cal. Aug. 6, 2021) (denying a lead-plaintiff motion by three public funds where, inter alia, the funds had “no pre-existing relationship,” “provide[d] no explanation why [they] ha[d] chosen to retain two separate law firms,” and another institutional investor had losses greater than those of any of the group members). As a result, hands of the lead plaintiff.” Cavanaugh, 306 F.3d at 734. Thus, “if the lead plaintiff has made a reasonable choice of counsel, the district court should generally defer to that choice.” Cohen v. U.S. Dist. Court for N. Dist. of California, 586 F.3d 703, 712 (9th Cir. 2009); see also Cavanaugh, 306 F.3d at 733 (stating that courts should not disturb the lead plaintiff’s selection of counsel unless his choice is “so irrational, or so tainted by self-dealing or conflict of interest, as to cast genuine and serious doubt on that plaintiff’s willingness or ability to perform the functions of lead plaintiff”). The Court finds that the Retirement Systems’ selected firm, Saxena White, is fit to serve as lead counsel in this case. According to the firm’s résumé, Saxena White has recovered hundreds of millions of dollars for plaintiffs in securities cases over the past decade. See Ret. Sys. Mot. at 14, Docket No. 30-6 (comprising the firm résumé). “Additionally, in this District, Saxena White achieved a settlement valued at $320 million in a derivative action on behalf of Wells Fargo & Company—one of the largest shareholder derivative settlements in history, including $240 million in cash.”18 Ret. Sys. Mot. at 14. And in appointing Saxena White co-lead counsel in a recent derivative suit, the Southern District of Ohio commended the firm for its “considerable track record[] of successfully prosecuting” securities actions and for its diverse makeup as “a federally- certified, minority- and women-owned firm.” Bloom v. Anderson, 2020 WL 6710429, at *8-9 (S.D. Ohio Nov. 16, 2020). The opposing parties, finally, have not attempted to show that Saxena White would be in any way inadequate to representing the proposed class. The Court therefore GRANTS the Retirement Systems’ motion to approve Saxena White as lead counsel and DENIES Sepulveda’s and Branca-Mollo’s corresponding motions. For the reasons given above, the Court GRANTS the plaintiffs’ motions to consolidate the related actions in this matter. It also GRANTS the Retirement Systems’ motion for appointment as lead plaintiffs and for approval of lead counsel and DENIES the corresponding motions of

18 The Retirement Systems also aver that Saxena White is “a federally certified woman- and Sepulveda and Branca-Mollo. Plaintiffs shall file a consolidated amended complaint within thirty (30) days from the date of this order. Plaintiffs’ counsel shall also submit a protocol for controlling fees and costs as this Court has implemented in, e.g., In re Carrier IQ Consumer Privacy Litigation, No. 12-md-02330-EMC, Docket No. 100 (N.D. Cal. July 16, 2012). This order disposes of Docket Nos. 22, 29, and 40. Dated: August 30, 2021 ______________________________________ EDWARD M. CHEN United States District Judge