LOUISIANA WHOLESALE DRUG CO., INC. v. SMITHKLINE BEECHAM CORPORATION

Opinion

Not for Publication

UNITED STATES DISTRICT COURT DISTRICT OF NEW JERSEY

In re: Lamictal Direct Purchaser Antitrust Litigation Civil Action No. 12-995

OPINION

John Michael Vazquez, U.S.D.J. This antitrust class action stems from a settlement in a patent lawsuit concerning a brand drug, Lamictal, and its generic competitor, lamotrigine. In the patent case, Defendant SmithKline Beecham Corporation d/b/a GlaxoSmithKline (“GSK”) sued Defendants Teva Pharmaceutical Industries LTD and its subsidiary Teva Pharmaceuticals USA, Inc. (collectively “Teva”). GSK and Teva settled the matter, with GSK promising not to launch an authorized generic version of Lamictal for a specified period. Plaintiffs in the current matter are purchasers of Lamictal and lamotrigine and claim that the agreement not to launch an authorized generic constituted an improper “reverse payment,” causing Plaintiffs to pay more than they would have if GSK had sold the authorized generic. But there is a twist. GSK claims that Plaintiffs were not harmed because GSK lowered the prices of Lamictal through a contracting strategy. And Teva adds that it learned of GSK’s contracting strategy and preemptively lowered the price of lamotrigine. This matter is on remand from the Third Circuit Court of Appeals’ decision in In re Lamictal Direct Purchaser Antitrust Litig., 957 F.3d 184 (3d Cir. 2020), which vacated this Court’s previous Opinion, D.E. 428, and Order, D.E. 429, granting Plaintiffs’ motion for class certification, D.E. 371.1 The Third Circuit instructed the Court to perform a rigorous analysis in determining whether to certify the class as to direct purchasers of lamotrigine.2 The Third Circuit further instructed the Court to address certain predicate questions before assessing Plaintiffs’ use of averages to prove antitrust injury as to each class member. Following remand, the parties filed supplemental briefs, D.E. 478 (“Br.”), D.E. 479 (“Opp.”), D.E. 483 (“Reply”). The Court then

held oral argument. D.E. 498. For the reasons that follow, the Court determines that Plaintiffs have not shown by a preponderance of the evidence that they can prove antitrust injury through common evidence as to generic-only purchasers. I. BACKGROUND The underlying patent suit was filed in April 2002 and involved litigation pursuant to the Drug Price Competition and Patent Term Restoration Act of 1984, commonly known as the Hatch- Waxman Act, 21 U.S.C. § 355, et seq., which was discussed in detail in the Court’s prior Opinions. See e.g., D.E. 105 at 2-4. GSK sued Teva, alleging that Teva’s generic drug, lamotrigine, infringed GSK’s patent for Lamictal. See D.E. 55 (“Class Compl.”) ¶ 13. Lamictal is used to treat epilepsy and bipolar disorder. On February 16, 2005, GSK and Teva settled the patent case. D.E. 373-3 at 2.3 As part of the settlement, GSK agreed that Teva could begin selling lamotrigine on July 22,

2008. D.E. 373-4 at 16. GSK further promised that it would refrain from launching its own competing authorized generic version of Lamictal (the “No-AG Promise”) until Lamictal’s patent expired on January 22, 2009. See id.; see also D.E. 406-2 at 10-11, ¶ 17. This No-AG promise is

1 The matter was transferred to the undersigned following the remand.

2 The proposed class includes direct purchasers of both Lamictal and generic lamotrigine. The Circuit’s opinion, however, only addressed the purchasers of the generic.

3 Unless otherwise specified, page numbers for exhibits reflect the Court’s electronic filing system’s generated page numbers. at the heart of the current dispute, with Plaintiffs contending that it constituted an improper reverse payment. Plaintiffs claim that absent the settlement agreement, either (1) Teva would have prevailed in the patent litigation, allowing for an earlier launch of generic lamotrigine tablets as well as triggering Teva’s 180-day generic exclusivity, or (2) Teva would have launched its generic version

of lamotrigine tablets “at risk” (i.e., before the patent case was decided) following the FDA’s final approval and the expiration of the 30-month statutory stay. Id. ¶ 26. Thus, Plaintiffs contend that absent the settlement, GSK would have faced price competition for its Lamictal products through Teva’s generic launch, and that absent the No-AG Promise, Teva’s generic drug would have faced pricing competition from GSK’s authorized generic. Id. ¶ 28. Plaintiffs continue that the lack of competition that resulted from the No-AG promise forced Plaintiffs to purchase both Lamictal and lamotrigine at artificially inflated prices. Id. On June 28, 2018, Plaintiffs moved to certify the following class: All persons or entities in the United States and its territories who purchased Lamictal Tablets directly from GSK, or who purchased a generic version of lamotrigine tablets directly from Teva, at any time during the Class Period from February 17, 2008 until January 22, 2009.

D.E. 372 at 3. As to the predominance requirement of Fed R. Civ. P. 23(b)(3), Plaintiffs argued that it was met because Plaintiffs could show antitrust impact through common evidence – “empirical economic research,” id. at 32, “documents, testimony, and forecasting documents regarding the effects of generic competition,” id. at 33, and data, id. at 34. Defendants countered that this case was different from other reverse-payment cases because GSK employed a “contracting strategy” (“Contracting Strategy”) for Lamictal. D.E. 406 at 5. GSK asserted that because “[m]any doctors were concerned that switching to a generic could reduce the drug’s effectiveness[,]” GSK believed that “the magnitude of [generic] erosion [would] not approach that seen in most other disease areas.” Id. Thus, GSK offered discounts up to 40% on Lamictal to certain customers to compete with Teva’s generic. Id. at 6. In exchange for GSK’s discount, the customers receiving discounts agreed to dispense “[Lamictal] as a generic.” Id. Defendants further claimed that Teva learned of

GSK’s Contracting Strategy, sued GSK for breach of the parties’ settlement agreement, and preemptively lowered the price for lamotrigine. Id. at 6-7. Therefore, Defendants contended that Plaintiffs could not use evidence about average changes in generic prices when a second generic enters the market to prove antitrust injury because doing so would ignore the reality that Teva faced competition – through the Contracting Strategy – from GSK’s discounted brand Lamictal. Id. at 20-22. Judge Walls rejected Defendants’ arguments and certified the class. D.E. 428 at 13- 15. On appeal, Defendants only challenged “the [d]istrict [c]ourt’s predominance finding,” id. at 190, “as to the members who purchased generic lamotrigine from Teva.” Lamictal, 957 F.3d at

195. The Third Circuit first reviewed a district judge’s duties in assessing a class certification motion, stating that a district court “must conduct a ‘rigorous analysis’ of the evidence and arguments presented” at class certification. Id. at 190-91 (citing In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 309 (3d Cir. 2008), as amended (Jan. 16, 2009)). The Circuit continued that (1) “the court must ‘find’ that the requirements of Rule 23 are met and any ‘[f]actual determinations supporting Rule 23 findings must be made by a preponderance of the evidence’”; (2) “the court must resolve all factual or legal disputes relevant to class certification, even if they overlap with the merits”; and (3) “the court must consider ‘all relevant evidence and arguments,’ including ‘expert testimony, whether offered by a party seeking class certification or by a party opposing it[.]’” Id. (citing Hydrogen Peroxide, 552 F.3d at 307). The Circuit concluded that if, after considering the foregoing, the district court “is convinced by a preponderance of the evidence that the plaintiffs’ claims are capable of common proof at trial, then the predominance requirement is satisfied.” Id. The Circuit framed the issue before it in the following terms:

With that in mind, we turn to the Direct Purchasers’4 claim. The injury element is at issue here. Recall that their theory of liability is that they suffered an antitrust injury because but for the reverse- settlement agreement, each would have paid less for lamotrigine than it actually did. This requires the Direct Purchasers to prove by a preponderance of the evidence that they could establish, through common proof at trial, facts supporting an antitrust injury, namely: 1) GSK would have launched an AG but for the reverse-settlement; and 2) as a result, all class members would have paid less for lamotrigine in this but-for world. If each individual class member could rely on this same proof to prove the elements of its claim, then the injury is capable of common proof at trial.

Id. at 192 (first emphasis added). The Third Circuit then ruled that “a more rigorous analysis” was needed to determine whether Plaintiffs’ expert’s reliance on averages was acceptable in light of GSK’s Contracting Strategy. The Third Circuit found that the Court erred when it “refused to ‘address the multi- leveled microeconomic analysis of what each Defendant would or would not have possibly done in the but-for world.” Id. at 193. Without that inquiry, the Third Circuit found that “it is impossible to determine whether the Contracting Strategy raised individual issues.” Id.

4 The Third Circuit defined Plaintiffs as “Direct Purchasers” meaning “companies that directly purchased brand Lamictal from GSK or lamotrigine from Teva.” Id. at 189. However, for purposes of the appeal, the Circuit continued that “GSK and Teva challenge only certification as to the members who purchased generic lamotrigine from Teva (hence any reference to the Direct Purchasers that follows is limited to those Direct Purchasers).” Id. at 190. As a result, the Court does the same on remand and focuses on Plaintiffs who purchased the generic from Teva. Specifically, the Third Circuit found that the Court erred in failing to resolve issues raised in the parties’ dueling expert reports. Plaintiffs’ expert (Dr. Russell Lamb), the Circuit noted, relied on the following: (1) economic literature showing that, on average, prices of generics are lower as more enter the market; (2) Teva’s own general pricing forecast tending to discount a generic by 50% without competition, but by 65% when facing an additional competitor; and (3) transaction-level sales data showing that the average actual price paid was consistent with these predictions.

Id. Conversely, the Circuit continued, Defendants’ expert (Dr. James Hughes) opined that an individualized inquiry was necessary to determine whether any particular member of the class suffered an actual injury. Id. The Third Circuit also highlighted that Defendants’ expert (1) criticized Lamb’s use of averages, (2) opined that Lamb committed meaningful error when he assumed an aggregate actual price because he failed to acknowledge that purchasers paid significantly different prices, and (3) criticized Lamb’s use of general forecasting documents rather than lamotrigine-specific prices. Id. The Circuit further noted that Hughes developed his own model based on lamotrigine-specific prices from Teva documents which demonstrated that “25 of the 33 generic-only purchasers likely paid the same or lower prices in the actual world under the Contracting Strategy than they would have paid had GSK launched an AG.” Id. The Third Circuit then ruled as follows: Here, the District Court abused its discretion when it assumed, absent a rigorous analysis, that averages are acceptable. As is clear from the dueling expert reports, the acceptability of averages depends largely on the answer to several factual predicates, most importantly: 1) whether the market is characterized by individual negotiations; 2) whether Teva preemptively lowered its pricing in response to the Contracting Strategy; and 3) whether and to what extent GSK, absent the settlement agreement, would or could have pursued both the Contracting Strategy and an AG. The Court did not resolve these factual disputes, which would have required it to weigh the competing evidence and make a prediction as to how they would play out at trial. Further, much of each expert’s analysis turned on his sources of evidence for pricing and discounting data, many of which were in tension. It was up to the District Court to scrutinize the evidence to determine what was credible and could be used in the expert analysis.

Id. at 194. The Third Circuit added that if the Court had assumed antitrust injury, the Court had misunderstood Defendants’ argument that the “prices were never inflated to begin with because Teva preemptively lowered its prices before launching; thus, some Direct Purchasers never suffered an overcharge.” Id. The Circuit concluded that the Court also appeared to have conflated damages with antitrust injury. Id. II. ANALYSIS The Court first reviews antitrust injury, then resolves the three questions posed by the Third Circuit, and concludes with an analysis of related arguments raised in the supplemental briefing. A. Antitrust Injury Because the Third Circuit focused on antitrust injury (or antitrust standing), the Court reviews the applicable legal standards. “[E]ven [if] a plaintiff has established Article III standing, antitrust standing remains as a prerequisite to suit, focus[ing] on the nature of the plaintiff’s alleged injury, [and] asking whether it is of the type that the antitrust statute was intended to forestall.” Hartig Drug Co. Inc. v. Senju Pharm. Co. Ltd., 836 F.3d 261, 269 (3d Cir. 2016) (internal quotations omitted). Antitrust standing, unlike Article III standing, is a “prudential limitation” that does not affect the Court’s subject-matter jurisdiction, but rather “prevents a plaintiff from recovering under the antitrust laws.” Ethypharm S.A. v. Abott Laby’s, 707 F.3d 223, 232 (3rd Cir. 2013); see also Sullivan v. DB Invs. Inc., 667 F.3d 273, 307 (3d Cir. 2011) (“[L]ack of antitrust standing affects a plaintiff’s ability to recover, but does not implicate the subject-matter jurisdiction of the court.”) (citing Gerlinger v. Amazon.com Inc., 526 F.3d 1253, 1256 (9th Cir. 2008))). Antitrust standing requires a plaintiff to “prove more than injury causally linked to an illegal presence in the market.” Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977). This inquiry looks to whether the plaintiff suffered an antitrust injury, meaning an “injury

of the type the antitrust laws were intended to prevent and that flows from that which makes defendants’ acts unlawful.” Id.5 In class actions, the Third Circuit has observed the following: In an antitrust class action, “impact often is critically important for the purpose of evaluating Rule 23(b)(3)’s predominance requirement because it is an element of the claim that may call for individual, as opposed to common, proof.” [Hydrogen Peroxide, 552 F.3d] at 311. A district court must thus undertake a “rigorous assessment of the available evidence and the method or methods by which plaintiffs propose to use the evidence to prove impact at trial.” Id. at 312. The class should only be certified “if such impact is plausible in theory [and] it is also susceptible to proof at trial through available evidence common to the class.” Id. at 325. This inquiry often involves an overlap into the merits. Id. at 324.

5 The Supreme Court in Associated General Contractors of California, Inc. v. California State Council of Carpenters articulated several factors to consider when determining whether a plaintiff has antitrust standing. 459 U.S. 519, 535-38 (1983). Since then, the Third Circuit has “organized those factors . . . into the following multifactor test”:

(1) the causal connection between the antitrust violation and the harm to the plaintiff and the intent by the defendant to cause that harm, with neither factor alone conferring standing; (2) whether the plaintiff’s alleged injury is of the type for which the antitrust laws were intended to provide redress; (3) the directness of the injury, which addresses the concerns that liberal application of standing principles might produce speculative claims; (4) the existence of more direct victims of the alleged antitrust violations; and (5) the potential for duplicative recovery or complex apportionment of damages.

Ethypharm S.A. France, 707 F.3d at 232-33 (quoting In re Lower Lake Erie Iron Ore Antitrust Litig., 998 F.2d 1144, 1165-66 (3d Cir. 1993)). In re Modafanil Antitrust Litig., 837 F.3d 238, 262-63 (3d Cir. 2016). To be clear, Plaintiffs are asserting a harm that would qualify as an antitrust injury. Plaintiffs are claiming that due to the No-AG Promise, they (as direct purchasers) paid higher prices because Teva’s lamotrigine did not face price competition from GSK’s authorized generic. This theory fits squarely within a cognizable antitrust injury.

One example of an insufficient antitrust impact was discussed in Brunswick Corp., 429 US. at 477. In that case, the plaintiffs operated bowling centers and alleged that the defendant violated the Clayton Act. The defendant was a large manufacturer of bowling equipment who had been acquiring bowling centers that were in default. Id. at 479-80. The plaintiffs contended that if the defendants had not purchased the defaulting centers, then the plaintiffs would have made more money because they would have faced less competition. Id. at 481. The issue before the Brunswick Court was “whether antitrust damages are available where the sole injury alleged is that competitors were continued in business, thereby denying respondents an anticipated increase in market shares.” Id. at 484. The Supreme Court determined that the plaintiffs suffered no antitrust

injury because the complained of harm – lost profits because of more competition – was the exact opposite of the harm that the antitrust laws were designed to guard against. Id. at 488. Other times, a plaintiff cannot establish antitrust injury because its relationship to the alleged antitrust conduct is too attenuated. For example, in Mid-West Paper Products Co. v. Continental Group, Inc., 596 F.2d 573, 580 (3d Cir. 1979), the plaintiff claimed that it was injured because it purchased bags from a competitor of the defendants and that the competitor was able to charge higher prices because of the defendants’ price fixing. The Third Circuit ruled that the plaintiff did not have antitrust standing because it was not a direct customer of the defendants. Id. at 583. By comparison, Plaintiffs here are direct purchasers of lamotrigine. Thus, the question before the Court is not whether Plaintiffs’ alleged harm falls within the parameters of antitrust injury; it clearly does. Instead, the question is whether Plaintiffs have shown that they can prove such harm to all class members through common evidence. B. The Third Circuit’s Predicate Questions

As noted, The Third Circuit ruled as follows: As is clear from the dueling expert reports, the acceptability of averages depends largely on the answer to several factual predicates, most importantly: 1) whether the market is characterized by individual negotiations; 2) whether Teva preemptively lowered its pricing in response to the Contracting Strategy; and 3) whether and to what extent GSK, absent the settlement agreement, would or could have pursued both the Contracting Strategy and an AG. The Court did not resolve these factual disputes, which would have required it to weigh the competing evidence and make a prediction as to how they would play out at trial. Further, much of each expert’s analysis turned on his sources of evidence for pricing and discounting data, many of which were in tension. It was up to the District Court to scrutinize the evidence to determine what was credible and could be used in the expert analysis.

Lamictal, 957 F.3d at 194 (emphasis added). These factual predicates all concern a common query – the “acceptability of averages” in evaluating classwide antitrust injury.6 1. Individualized Negotiations in the Market The first inquiry from the Third Circuit is “whether the market is characterized by individual negotiations.” Id. at 194. Plaintiffs’ expert defined the relevant market as that for “lamotrigine tablets . . . [in] the United States.” D.E. 373 ¶ 47. Accordingly, the question before

6 Of note, Plaintiffs represent that they will no longer rely on Dr. Lamb’s sensitivity analysis in support of class certification. Br. at 6. The Third Circuit described Dr. Lamb’s sensitivity analysis as follows: “While the parties dispute whether GSK would have used both strategies—launching an AG and engaging in the Contracting Strategy—simultaneously, Lamb conducted a ‘sensitivity analysis’ as part of his model purporting to show that, either way, the price of lamotrigine would have been lower absent the settlement agreement.” Lamictal, 957 F.3d at 193 n.4. Accordingly, the Court does not consider the analysis here. the Court is whether the market for lamotrigine tablets in the United States is characterized by individual negotiations. For reasons unknown, Plaintiffs do not address this issue in the supplemental briefing. See Br. at 2 (addressing three questions, none of which discuss whether the market is characterized by individual negotiations); see also Opp. at 3 n.1 (“Plaintiffs do not even dispute that Teva

individually negotiated prices with each lamotrigine customer.”). Plaintiffs have not put forward argument as to whether the market for lamotrigine in the United States was characterized by individual negotiations. Defendants proofs as to individualized negotiations focus on the Contracting Strategy. See D.E. 406 at 20-22. However, based on the evidence Defendants submitted, Teva appears to have contemplated and implemented different lamotrigine prices for different groups of customers. For example, Defendants’ expert divided Teva’s customers into categories when discussing Teva’s price reductions in response to the Contracting Strategy. Dr. Hughes found that 13 customers received discounts of 20.8%, 2 received discounts of 16.7%, and 10 received discounts of 9.3%. Opp. at 11. These categories, or groups, of customers are discussed

in more detail in the next section. As a result, the Court concludes that Teva’s prices for lamotrigine varied by specific groups of customers rather than each customer individually. 2. Teva Preemptively Lowering its Generic Prices The next issue on remand is “whether Teva preemptively lowered its pricing in response to the Contracting Strategy.” Lamictal, 957 F.3d at 194. The Third Circuit commented that this question “is a factual matter hotly contested by the parties” and instructed this Court “to resolve that dispute by a preponderance of the evidence.” Id. On remand, Plaintiffs make alternative arguments. First, Plaintiffs assert that Teva did not preemptively lower its prices in response to the Contracting Strategy. Br. at 9-14. Second, Plaintiffs indicate that even if Teva did preemptively lower its prices, Plaintiffs can show by common evidence that Teva would have lowered it prices further if it faced competition from an authorized generic rather than just the contracting strategy. Br. at 15-19. The Court is somewhat surprised that this issue is hotly contested as it seems susceptible to a straightforward answer because either Teva did, or Teva did not, preemptively lower its

lamotrigine prices in response to GSK’s Contracting Strategy. From the best that the Court can discern, it appears that part of the difficulty may be attributed to Plaintiffs’ not obtaining fact discovery on this issue. It appears that Plaintiffs were aware of GSK’s Contracting Strategy (although Plaintiffs may not have known how GSK referred to the strategy). Plaintiffs also knew that Teva was aware of the Contracting Strategy at the time it launched lamotrigine, July 22, 2008.7 See e.g., Class Compl. ¶ 80; see also id. ¶ 87 (discussing Teva’s allegation “in the subsequent litigation” that “GSK implemented a scheme to slow Teva’s market penetration for its generic version of Lamictal Tablets.”)). However, while Plaintiffs may have been aware of the Contracting Strategy, it appears that Plaintiffs did not learn during discovery that Teva allegedly lowered its

lamotrigine prices as a result of the strategy. Defendants rely on a declaration by Maureen Cavanaugh; Cavanaugh was deposed but did not answer questions as to Teva’s purported preemptive pricing response to the Contracting Strategy. In a declaration dated February 23, 2018, Cavanaugh indicated that Teva learned of the Contracting Strategy and decided to lower its prices for lamotrigine. D.E. 476-2 (“Cavanaugh

7 The day after the lamotrigine launch, on July 23, 2008, Teva sued GSK for breach of their settlement agreement because of the Contracting Strategy. Teva Pharm. Indus. Ltd. v. Smithkline Beecham Corp., Civ. No. 08-3706, 2009 WL 1687457 (D.N.J. June 16, 2009). Judge Cavanaugh, who was assigned to the case, described the matter as follows: “Just prior to Teva launching its generic, GSK approached various pharmacies, group purchasing organizations, and long-term care facilities and proposed that they purchase and distribute GSK’s Lamictal at a generic product price which would be eligible for generic product reimbursement.” Id. at *2. Declaration” or “Cavanaugh Decl.”). In 2008, Cavanaugh was Teva’s Director of Sales & Marketing and then Senior Director of Marketing. Cavanaugh Decl. ¶ 2. Cavanaugh was involved, along with “other employees” in preparing for Teva’s July 22, 2008 launch of lamotrigine. Id. at ¶ 4. Cavanaugh indicated that “[d]uring this time” (while preparing for the launch), “Teva learned that GSK had begun offering discounts to customers purchasing Lamictal tablets.” Id. at ¶ 5. As

a result, Cavanaugh continued, “Teva responded to GSK’s discounts by lowering the initial price that Teva would charge purchasers for lamotrigine tablets upon launch.” Id. at ¶ 6. In connection with the launch, Cavanaugh and another Teva employee, Kevin Galownia, built a spreadsheet with several tabs, including the “Original Pricing” tab and the “Updated Pricing” tab. Id. at ¶ 7. The Original Pricing tab reflected prices that Cavanaugh8 thought Teva should charge for lamotrigine before learning of the Contracting Strategy. Id. at ¶ 8. The Updated Pricing tab indicated prices that Cavanaugh thought Teva should charge after learning of the Contracting Strategy. Id. at ¶ 9. The prices in the Updated Pricing tab are “generally lower” when compared to those in the Original Pricing tab. Id. To be clear, it does not appear that the prices in either the

Original Pricing tab or the Updated Pricing tab, at least according to Cavanaugh, were the prices that Teva had actually decided to charge for lamotrigine. Cavanaugh added that “[w]e revised contract prices again shortly before Teva’s July 22, 2008 launch of lamotrigine tablets.” Id. at ¶ 10. Cavanaugh does not indicate how prices were revised (either higher or lower). Cavanaugh also does not indicate whether some, or all, prices were revised. Cavanaugh then pointed to a “a spreadsheet that compiled the initial orders that Teva’s customers placed for lamotrigine and the prices Teva charged those purchasers,” which

8 Cavanaugh stated that the tab “reflects the contract prices that we thought Teva should charge[.]” Id. at ¶ 7 (emphasis added). However, Cavanaugh did not indicate who the “we” consisted of. It is not clear if she meant Galownia or a broader group of people. Teva produced under the Bates Number TEVA00022998. Id. Cavanaugh stated the “Go Forward Pricing” tab of the spreadsheet summarized her analysis of the initial orders. Id. Cavanaugh concluded that “[t]he contract prices listed in the ‘Go Forward Pricing’ tab . . . are lower than those in the ‘Original Pricing’ tab . . . because Teva lowered its contract prices for lamotrigine tablets upon launch in response to GSK’s discounts on Lamictal Tablets.” Id. at ¶ 11.

Plaintiffs are highly critical of Cavanaugh’s Declaration. Plaintiffs first indicate that during her deposition, Cavanaugh responded, “I don’t recall” when asked if she had found any lamotrigine forecasts after searching all of her files. Br. at 11. Plaintiffs continue that Teva’s preemptive pricing argument is “a defense that appeared only after the close of discovery.” Br. at 11 n.17. Defendants respond that Cavanaugh did not testify about “Teva’s actual pricing in response to the Contracting Strategy at her deposition because Plaintiffs’ counsel inexplicably failed to ask her about it.” Opp. at 9. Assuming, without deciding, the accuracy of Plaintiffs’ arguments, Plaintiffs do not request any relief from the Court. Plaintiffs do not make a motion to reopen discovery to take another

deposition of Cavanaugh. Plaintiffs also do not provide any evidence showing that Defendants’ misled Plaintiffs during discovery as to Teva’s preemptive pricing strategy. And Plaintiffs do not invoke the “sham affidavit doctrine.” Baer v. Chase, 392 F.3d 609, 623-24 (3d Cir. 2004) (discussing the doctrine as to a motion for summary judgment). In fact, in the initial briefing before Judge Walls, Defendants similarly relied on Cavanaugh’s Declaration and supporting documents. D.E. 406 at 11-12. Plaintiffs did not challenge Cavanaugh’s Declaration or the associated pricing tabs, instead asserting in a footnote that information constituted “common evidence[.]” D.E. 419 at 13 n.13. As a result, the Court considers the Cavanaugh Declaration.9 Plaintiffs next assert that contemporaneous email evidence shows that Cavanaugh’s Declaration is incorrect. Plaintiffs submit that this evidence shows Teva had planned to sell its lamotrigine at 50% discount off the Lamictal WAC (“wholesale acquisition cost”) before learning

of the Contracting Strategy. Br. at 9. Plaintiffs assert that a July 9, 2008 email sent to Cavanaugh indicates that Teva’s generic price would be the same, 50% of WAC. Br. at 10. The email’s subject line is “Re: Latest IMS on Lamotrigine Tabs” and it does indicate “Price 50% of Brand WAC[.]” D.E. 421 at 2. At the same time, Cavanaugh did not respond to the email, so while the communication has some probative value it is not dispositive. Plaintiffs do not provide evidence that Cavanaugh expressly agreed with the contents of the email. Another unanswered question is what 50% of WAC would have amounted to in light of the Contracting Strategy; presumably the WAC would have been lower than it would have been without the strategy because the price of Lamictal was reduced.

Defendants also point to Galownia’s deposition testimony, Opp. at 17, in which he testified that Teva never intended to price lamotrigine at 50% of brand WAC. Instead, Galownia stated that Teva had the following pricing strategy based on customer categories,10 stated at a percentage

9 Moreover, even if Plaintiffs were not aware that Teva was claiming a preemptive price reduction in response to the Contracting Strategy, the Court would anticipate that Plaintiffs would have sought discovery that would have revealed such information. As noted, due to the 2008 litigation between Teva and GSK, Plaintiffs knew that Teva had been aware of the Contracting Strategy before Teva launched lamotrigine. Plaintiffs could have followed up with basic inquiries: when did Teva learn of the Contracting Strategy? What was Teva’s understanding of the Contracting Strategy (particularly as to the pricing of Lamictal)? And what, if any actions, did Teva take once it learned of the Contracting Strategy (particularly as to the pricing of lamotrigine)?

10 As to the first predicate question, this evidence also indicates Teva contemplated different prices based on categories of purchasers of WAC, D.E. 479-2 at 92-93, Tr. 159:6- 162:11. Plaintiffs do not address this evidence in their reply. Plaintiffs also refer to a Teva earnings conference call on November 6, 2008, several months after the launch of lamotrigine. D.E. 500. During the call, Teva’s President and CEO, Shlomo Yanai, was asked whether “there was anything going on with GSK during the launch that

maybe clipped the pricing?” D.E. 500-1 at 9. While acknowledging that GSK tried to do something “creative,” Yanai responded that GSK’s effort “was largely unsuccessful and really didn’t impact price all that much. Remember, Lamotrigine is our own API [active pharmaceutical ingredient]. It’s a relatively high margin product so I think there was no real issue.” Id. These comments do appear to undercut Defendants’ position, but, again, Plaintiffs do not provide evidence to explain the ambiguous or inexact words/terms such as “largely,” “all that much,” and “relatively.” Before turning to the remaining arguments as to Teva’s preemptive pricing strategy, the Court notes that there are numerous additional pieces of information that would have been helpful

to the Court’s analysis. For example, evidence as to whether the Contracting Strategy resulted in GSK’s Lamictal pricing to be equivalent to that of an authorized generic. According to GSK, due to the uniqueness of Lamictal, it believed that physicians would not be inclined to switch patients, who were stable on Lamictal, from the brand to the generic. Opp. at 5. Of course, such a notion begs the question – why engage in the Contracting Strategy at all if physicians were not going to switch from Lamictal for such patients? At a minimum, such a belief would seemingly indicate that GSK never believed that its Contracting Strategy prices had to be the same as those for an authorized generic because of GSK’s basic premise – physicians were unlikely to switch from Lamictal for patients who were stable. And, more importantly, once Teva learned11 of the Contracting Strategy before it launched lamotrigine, did Teva believe that the Contracting Strategy would result in GSK offering prices that were commensurate with those of an authorized generic. Because once Teva learned of the Contracting Strategy, it had no financial incentive to lower its prices for lamotrigine to the same level as it would have had it been competing with an authorized generic unless Teva believed that

the Contracting Strategy forced it do so. Turning to the parties’ remaining arguments, Plaintiffs contend that Teva’s internal documents show that the prices Teva charged for generic lamotrigine were consistent with the prices Teva projected years prior to the launch. Br. at 10. In support, Plaintiffs cite to testimony from Dr. Lamb that in a “June 2006 Teva sales forecast for its lamotrigine tablets, Teva assumed that it would charge a price that was 50 percent of the brand Lamictal price for the first six months when [Teva] was the sole manufacturer.” D.E. 373-1 at 36, ¶ 65. Although Dr. Lamb claims this statement is supported by an “Exhibit 1172” to the “Galownia Deposition” the Court is unable to locate this exhibit and it is unclear whether Plaintiffs have provided it to the Court. Id. Plaintiffs

also rely on testimony from Dr. Lamb as to the July 9, 2008 email to Cavanaugh (discussed above) that forecasted generic Lamotrigine would be priced at 50 percent of brand WAC during Teva’s 180-day exclusivity period. Id. at 64, ¶ 112. However, even accepting Dr. Lamb’s analysis as accurate, it does not answer a fundamental question, that is, what was Lamictal’s WAC before and after the Contracting Strategy. It appears unrefuted that due to the Contracting Strategy, GSK

11 It is entirely unclear to the Court when or how Teva first learned of the Contracting Strategy. Obviously, Teva learned of the strategy before it sued GSK on July 23, 2008. Other than that, the most specific information is in Cavanaugh’s declaration, which indicates that she “and others at Teva” learned of the Contracting Strategy “[d]uring the first half of 2008.” Cavanaugh Decl. at ¶¶ 3, 5. lowered the price of Lamictal (up to 40%). Thus, even if Teva still priced lamotrigine at 50% of WAC, if the WAC was lower, then Teva charged less for lamotrigine. A review of the Original Pricing tab and the Updated Pricing tab reveals that those documents corroborate Cavanaugh’s Declaration.12 For example, in the Original Pricing tab,

13 D.E. 479-2 at 70.

Id. Original pricing for other amounts and doses of generic lamotrigine followed a similar pattern with

Id. The Updated Pricing tab reflects updated contract prices for “25mg 100s” of generic

lamotrigine as follows: (1) (2) (3) and (4) D.E. 479-2 at 75. Supervalu Inc., was projected to pay pricing similar to Id. Updated pricing for other amounts and doses of generic lamotrigine followed a

12 Returning to the first predicate question as to individualized negotiations, these tabs also reveal that Teva contemplated different prices based on categories of purchasers.

13 The other comparable prices that the Court refers to also apply to the same quantity and dose. similar pattern with paying the highest contract prices; paying the lowest contract prices; and paying an amount in between. Id. Thus, the projected contract prices on the Updated Pricing tab were lower for a majority of

the purchasers than the projected prices in the Original Pricing tab. It appears that Teva was projecting fairly significant discounts to However, were not projected to receive discounts (as both the Original Pricing tab and the Updated Pricing tab reflected a price of $166.38 for “25mg 100s” of lamotrigine). Cavanaugh’s declaration along with the Original Pricing tab and the Updated Pricing tab indicate that when Teva learned of GSK’s Contracting Strategy, Teva initially projected that it would lower prices for most of its customers. The remaining question is whether and to what extent the updated pricing was actually implemented.

The parties’ dispute centers on their disparate interpretations of the Go Forward Pricing tab in the spreadsheet produced under Bates Number TEVA00022998. See D.E. 479-2 at 79.14 As noted, Cavanaugh indicated that TEVA00022998 is “a spreadsheet that compiled the initial orders that Teva’s customers placed for lamotrigine and the prices Teva charged those purchasers.” Cavanaugh Decl. at ¶ 10. Cavanaugh further explained that the Go Forward Pricing tab in the spreadsheet contained “the summary of [her] analysis.” Id.

14 Due to the size of TEVA00022998, Defendant provided a copy to the Court via USB flash drive and Defendants’ Exhibit 12 was provided as a slip page. For the sake of clarity, the Court will cite to TEVA00022998 using the designation “D.E. 479-2 at 79” and will reference appropriate tab names and row and column numbers and titles where applicable. Plaintiffs first argue that some prices on the Go Forward Pricing tab are higher than the prices in the Updated Pricing tab. Br. at 12. Defendants counter that it is not surprising that some prices are higher because the “Updated Pricing tab was Teva’s initial internal response” to the Contracting Strategy whereas the Go Forward Pricing tab reflects prices customers actually paid, which “played out differently in some instances.” Opp. at 10. The Court agrees with Defendants

that just because some purchasers did not ultimately receive a discount of the magnitude projected in the Updated Pricing tab, it does not mean that discounts did not occur or otherwise impact the accuracy of the Go Forward Pricing tab. Plaintiffs next argue that the “‘Go Forward’ net ‘prices’ are flatly inconsistent with the net prices Teva actually charges as shown in its own contemporaneous transactional sales data.” Br. at 13 (emphasis added). Defendants respond that “[n]either Ms. Cavanaugh’s testimony nor the spreadsheet purports to reflect actual net prices[.]” Opp. at 11. Instead, Defendants say the spreadsheets reflect the gross (or “contract”) price charged. Id. Gross prices (or “contract prices”) appear to be the price set to be paid under the relevant contract, whereas the net price reflects the

gross price less “all rebates and discounts ultimately provided.” Id.; see also D.E. 421 at 67, n. 3 (“Net prices are gross prices net of returns, credits, chargebacks, and rebates.”). Plaintiffs appear to admit that they are attempting to compare gross prices with net prices, see Br. at 14, but still argue that “whether Class members would have paid lower net prices is what matters.” Id. The Court agrees with Defendants that the spreadsheet cannot be attacked as inaccurate for failing to reflect prices it does not purport to record. See Cavanaugh Decl. ¶ 11 (“The contract prices listed in the ‘Go Forward Pricing’ tab . . . .” (emphasis added)). Plaintiffs next assert, D.E. 476-1 (“Coslett Declaration” or “Coslett Decl.”), that the prices reflected in the Go Forward Pricing tab do not reflect the gross prices actually paid by some of Teva’s customers. See id.; see also Br. at 14. Plaintiffs argue that prices attributed to certain categories of customers in the Go Forward Pricing tab do not match the prices actually paid by those customers. Id. Plaintiffs contend that these discrepancies “impeach Ms. Cavanaugh’s declaration and expose it as inaccurate and unreliable.” Id. at 3, ¶ 10. Defendants respond that Plaintiffs incorrectly ascribed the prices associated with “other small distributors” and “other small

retailers” on the Go Forward Pricing tab to 18 of the 21 customers listed in the Coslett Declaration. Id. at 11-12. As to those customers, Defendants contend that “Sheet 1” of the spreadsheet TEVA00022998 reflects the correct prices paid. Id. at 13. Defendants also refute these discrepancies through an exhibit they compiled (“Coslett table replication”) and through a declaration. D.E. 479-1 at 4 ¶ 19 (arguing the analysis in Exhibit 16 “shows that TEVA00022998 is fully consistent with the Teva transactional data.”); see also D.E. 479-3 at 2. Specifically, Defendants explain that TEVA0002298 “contains nine tabs, including the Go Forward tab – the only one Plaintiffs address. The Go Forward tab summarizes the date and analysis on the other eight tabs. Unlike the Go Forward tab, the other tabs expressly list the prices charged to every customer, including those Plaintiffs had pricing discrepancies.” Opp. at 13 n. 8.15

The Go Forward Pricing tab does not list every purchaser of lamotrigine from Teva. After listing several purchasers, the tab also includes categories of “other small distributors” and “other small retailers.” See D.E. 479-2 at 79. The prices reflected on the Go Forward Pricing tab for the

15 In reply, Plaintiffs claim that this explanation undermines Cavanaugh’s Declaration and argue that the Court should therefore exclude it. Reply at 8-9. The Court disagrees. At paragraph 10 of her declaration, Cavanaugh explained that “[w]e revised contract prices again shortly before Teva’s July 22, 2008 launch of lamotrigine tablets. Two months after the launch, I prepared a spreadsheet that compiled the initial orders that Teva’s customers placed for lamotrigine tablets and the prices Teva charged those purchasers. I understand that Teva has produced that spreadsheet under Bates Number TEVA00022998. That tab labeled ‘Go Forward Pricing’ contains the summary of my analysis.” Cavanaugh Decl. ¶ 10 (emphasis added). purchasers who fall into those “other” categories do not match the prices actually paid in the underlying transaction data. Defendants seemingly admit that the Go Forward Pricing tab provides an inaccurate summary as to the “Contract Price” for “other small distributors” and “other small retailers.”16 See Opp. at 13. For example, the Go Forward Pricing tab lists as the “Go Forward” price for a 100-count bottle of 100mg tablets of generic lamotrigine for

D.E. 479-2 at 79. Through their “Coslett table replication,” Defendants admit that some of the customers who arguably fall into those categories paid higher prices when compared to the Go Forward Pricing tab. See D.E. 479- 3 at 2. Although the Go Forward Pricing tab apparently represented that for a 100-count bottle of 100mg tablets of lamotrigine, Defendants concede that Id. Likewise, although the Go Forward Pricing tab apparently represented or the same quantity and dose, Defendants concede that Id. However, Defendants are correct that the contract prices reflected in “product price”

column of “Sheet 1” – the second tab of TEVA00022998 – match the prices listed in the “Actual Gross Price in Teva’s Actual Sales Data.”17 See D.E. 479-2 at 79. And a review of the correct pricing data, which Plaintiffs concede is accurate, shows that Teva did preemptively lower its pricing in response to the Contracting Strategy for most of its customers. The Original Pricing

16 However, it should be noted that Defendants do not concede that the purchasers for whom Plaintiffs claim the Go Forward Pricing tab is inaccurate fall into the “other” categories. Opp. at 12-13.

17 It appears that in limited instances purchased lamotrigine from Teva at a higher price than reflected on “Sheet 1” of TEVA00022998. See D.E. 479-3 at 2. However, Sheet 1 does accurately report the other prices those purchasers paid for lamotrigine. tab shows that, prior to learning of GSK’s Contracting Strategy, Teva projected that for a 100-count bottle of 100 milligram tablets of generic lamotrigine and that or the same. See D.E. 479-2 at 720. As with the Go Forward Pricing tab shows that many of the customers listed received discounts

for lamotrigine when compared to the prices they were projected to pay before Teva learned of GSK’s Contracting Strategy. Compare D.E. 479-2 at 70 (projecting that or a 100-count bottle of 100 milligram generic lamotrigine tablets) with D.E. 479-2 at 79 (reflecting that per 100-count bottle of 100 milligram generic lamotrigine tablets). Thus, the pricing data that Plaintiffs’ concede is accurate, see D.E. 476-1 at 4, reveals that most potential class members received discounts for their purchases of generic lamotrigine from Teva when compared to the prices that they were projected to pay before Teva learned of GSK’s Contracting Strategy. Compare D.E. 479-2 at 70 with Coslett Decl. at 4. Plaintiffs also assert that the prices on the Go Forward Pricing tab listed for

are inaccurate. Defendants respond that Plaintiffs are incorrectly comparing the “indirect” contract prices attributable to those customers with the transaction data without accounting for amounts those customers were able to “chargeback” to Teva. Opp. at 13-14 (“[B]ecause the price initially paid to Teva is the higher direct contract price, that is the gross price that is directly observable in the transactional data. The indirect contract prices do not match the gross prices in the transactional data (nor should they)

18 Plaintiffs concede that is one of the small retailers.” D.E. 476-1 at 8, n.3.

19 “In accordance with Dr. Hughes’ methodology . . . . [is] assigned to the ‘other small distributor’ category.” D.E. 476-1 at 8, n.3. unless one takes chargebacks into account.”). Defendants explain that Teva entered into two kinds of contracts with wholesale purchasers: direct and indirect. Id. at 13. Defendants continue that direct contracts set the upfront price for a wholesaler’s purchase and that indirect contracts set “a lower price Teva agreed to honor via chargebacks when the wholesaler resells the product.” Id. at 14. Defendants further explain that for wholesalers it is the “indirect price that is . . . listed on the

Go Forward tab as the contract price for such customers . . . [but] because the price initially paid to Teva is the higher direct contract price, that is the gross price that is directly observable in the transactional data.” Id. Plaintiffs do not take issue with Defendants’ explanation of the difference between direct and indirect contracts. See Reply at 7-10. Sheet 1 confirms Defendants’ explanation. See D.E. 479-2 at 79, Sheet 1. For example, Sheet 1 has 8 entries for Id. at Rows 21- 28. Under the column titled “Dir/Ind Flag” on Sheet 1, there are four entries listed as “direct” contracts and four entries listed as “indirect” contracts. Id. at Rows 25-28. As to the indirect contract price for a 100-count bottle of 100 milligram generic lamotrigine tablets, Sheet 1 reflects

that Id. at Row 21. This matches the price listed on the Go Forward Pricing tab. See D.E. 479-2 at 79, Go Forward Pricing tab. Sheet 1 also shows that as a direct contract price for a 100-count bottle of 100 milligram generic lamotrigine tablets. See D.E. 479-2 at 79, Sheet 1 at 25. This matches the amount Plaintiffs claim is reflected in Teva’s transactional data. Compare id. with Coslett Declaration at 4. Similarly, as to the direct contract prices reflected in Sheet 1 match the amounts Plaintiffs claim are reflected in Teva’s transactional data, compare D.E. 479-2, at 79, Sheet 1 at Rows 111- 14 with Coslett Declaration at 4, and the indirect contract prices reflected in Sheet 1 also match those listed on the Go Forward Pricing tab. See D.E. 479-2 at 79. As to the direct contract prices reflected in Sheet 1 match the amounts Plaintiffs claim are reflected in Teva’s transactional data, see D.E. 479-2 at 79, Rows 259-262. And four of the eight indirect contract prices reflected in Sheet 1 also match those listed on the Go Forward Pricing tab. Id. at Rows 255- 58. However, there are four “Multi-Source 661340-1” indirect contracts with prices that do not match those listed on the Go Forward Pricing tab. Id. at Rows 251-54; see also D.E. 479-3 at 2,

n. 2 ( . Despite those four inconsistencies, the Court is unaware of any other discrepancy between the prices Teva claims were charged in TEVA00022998 and the prices that Plaintiffs have derived from the transactional data. Although the Court wishes that the proofs would have been more specific, based on the current record, the Court finds by a preponderance of the evidence that Teva did preemptively lower of its prices for lamotrigine for most of its customers in response to the Contracting Strategy. 3. GSK’s Pursuit of Both the Contracting Strategy and an Authorized Generic Absent the Settlement Agreement

The final issue from the Circuit is “whether and to what extent GSK, absent the settlement agreement, would or could have pursued both the Contracting Strategy and an AG.” Lamictal, 957 F.3d at 194. Plaintiffs contend that absent the No-AG Promise, GSK would have simultaneously pursued both the Contracting Strategy and distribution of an authorized generic. Br. at 7. Plaintiffs further argue that the Contracting Strategy and launching an authorized generic would have been complementary strategies. Id. Plaintiffs state that “[n]o GSK witness even implied that the brand contracting strategy was a reason for GSK not to launch an [authorized generic].” Id. at 8-9. Defendants counter that it would have been illogical for GSK to pursue both an authorized generic and the Contracting Strategy concurrently because, under the Contracting Strategy, pharmacies were required to dispense Lamictal as a generic so GSK would have been illogically competing with itself had it launched an authorized generic. Opp. at 6-7. Plaintiffs rely on Dr. Lamb’s Revised Expert Reply Report. Br. at 7. Dr. Lamb stated that David Ballesteros of GSK “testified that the decision not to launch an authorized generic lamotrigine tablet was because GSK was prevented from doing so under [the No-AG Agreement]

not because it had chosen to pursue a brand contracting strategy instead.”20 D.E. 373-1 at 67, ¶ 104. Dr. Lamb also indicated that Rick Proctor of GSK testified to the same effect. Id. at 67-68, ¶ 105. Dr. Lamb’s report further referenced testimony of Diane Tulp of GSK that GSK was considering discounting brand Lamictal prior to (or just after) the No-AG Agreement. Id. at 68, ¶ 106. Based on this testimony, Dr. Lamb opined that “GSK would have launched an authorized generic and engaged in brand contracting, and it would have been economically rational for GSK to have engaged in brand contracting regardless of whether GSK launched an authorized generic. Id. ¶ 107, see also id. at 69 ¶ 108 (“[I]t is appropriate to assume that in the but-for world, GSK would have launched an authorized generic lamotrigine tablet at the same time Teva launched its

generic lamotrigine tablet in the but-for world[.]”). Plaintiffs also rely on Luis Molina, an expert with professional experience within the pharmaceutical industry. See D.E. 374-4 at 35. Molina indicated that absent the No-AG Promise,

20 The Court notes that Dr. Lamb’s analysis based on the testimony of GSK representatives does not appear to be within the purview of expert testimony. That is not to say the issue – whether GSK would have pursued both the Contracting Strategy and the launch of an authorized generic – does not raise topics that may be appropriate for expert analysis. For example, an expert apparently could opine whether GSK had the capacity to do both or an expert could seemingly provide insight as to whether it was economically rational to pursue both strategies simultaneously. Nevertheless, statements by GSK representatives are evidence, however, and it appears that Plaintiffs’ counsel could have argued the import of such evidence. Because such statements are evidence, and because Plaintiffs’ counsel would have been free to present such evidence to the Court, the Court considers it at this time. “GSK would have been ready, willing, and incentivized to simultaneously launch an authorized generic version of Lamictal tablets, and would have done so.” D.E. 374-4 at 6, ¶ 11. Molina continued that “GSK would have been able to launch an AG because there are few regulatory hurdles to doing so, and it had already selected its third-party distributor of the AG.” Id. at 10, ¶ 23. Molina added that “based on its previous AG launches, GSK had determined that it made

financial sense to launch an AG and planned to do so.” Id. Molina added that he did “not view the Contracting Strategy as an ‘alternative’ form of competition. Launching an authorized generic and engaging in the Contracting Strategy to try to maximize brand sales are not mutually exclusive.” D.E. 375-1 at 19 ¶ 24. Molina explained that the strategies are “complimentary” because “[a]n authorized generic allows the branded company to capture revenues from sales that would otherwise go to the generic company. The Contracting Strategy targets revenues from retained brand sales.” Id. Plaintiffs further claim that Defendants’ experts admitted that “brand contracting and an AG are complementary, not mutually exclusive, and so would have been deployed

simultaneously.” Br. at 7-8. Plaintiffs cite the deposition testimony of Dr. Edward A. Snyder in support. However, the testimony cited by Plaintiffs does not indicate what Plaintiffs claim it does. Plaintiffs’ counsel asked Dr. Snyder: Did you see any evidence that GSK would have preferred to do or sell an authorized generic version of Lamictal tablets once Teva entered the market instead of the contracting strategy and would not have done the contracting strategy if it weren’t for the promise it made to Teva not to launch an authorized generic?

D.E. 421 at 6, Tr. 57:23-58:6. Dr. Snyder responded: I have not seen evidence that related to all of the comparative preferred, instead of, and would have questions . . . I’ve seen things in the record . . . as to issues related to whether GSK wanted to or would have or how it evaluated costs and benefits, risks and opportunities, with an authorized generic. I saw a multitude of indications that this was in general something that required analysis . . . I saw additional analyses that were more specific to Lamictal . . . this was an analysis that was important and complicated and had not been done. I did not see anything that definitively said we are going to do this or we’re going to do that.

Id. at 8, Tr. 59: 3-6; 59:9-60:6. Plaintiffs also cite to the testimony of GSK’s employee David Ballesteros. Br. at 8, n.12. In response to being asked whether “the net sales proceeds realized by the authorized generic would be some share of the sales that would otherwise go to the generics who filed ANDAs and entered the market,” D.E. 476-8 at 4, Tr. 343:12-15, Ballesteros responded, “Perhaps, yes.” Id. at 4, Tr. 343:18. This answer is certainly not definitive. Plaintiffs also rely on the testimony of former GSK employee Diane Tulp. Tulp testified that the Contracting Strategy was “an opportunity . . . to capture some of those patients” who did not want to switch from Lamictal to generic lamotrigine out of fear of breakthrough seizures. D.E. 376-4 at 41, Tr. 157:20-25. Tulp’s response does not address the issue before the Court. Defendants contend that the evidence cited by Plaintiffs only shows that GSK could have launched an authorized generic along with the Contracting Strategy, not that it would have. Opp. at 5. Defendants continue that launching an authorized generic would have undermined the Contracting Strategy. Id. Defendants first point to a GSK document entitled “Global Commercial Plan: Compound/Product: Lamictal™ (IR and XR), Indications: Epilepsy & Neuropathic Pain (Painful Diabetic Neuropathy, Post-Herpetic Neuralgia), dated May 2003 and approved on June 2, 2003.” Id. The document provides that “there are nine critical success factors which will be addressed to achieve [GSK’s] objectives.” D.E. 479-2 at 4. One of the factors is the “Risk/Benefit assessment of any potential third-party agreements to supply Lamotrigine to generic houses to minimize the risk that this could seriously undermine the entire defence [sic] strategy.” Id. Later, the document notes that “[p]hysicians are extremely resistant to switching patients considered to be controlled on their current [antiepileptic drug] treatment regimen.” Id. at 6. The document continues that “[d]ue to the increased level of generic competition and the lack of a narrow therapeutic classification for the newer AEDs,21 it is believed that generic erosion of current

branded AEDs [including brand Lamictal] will occur more in the future than has been the case with older AEDs.” Id. at 6-7. The document adds that “the magnitude of this erosion will not approach that seen in most other diseases areas.” Id. at 7. The document does not necessarily support the conclusion that GSK believed that the launch of an authorized generic would have undermined its Contracting Strategy. Instead, it appears that GSK believed that physicians would be less likely to switch a patient effectively controlled on Lamictal. In fact, GSK believed that generic use would extend “to new patients or those who have yet to achieve adequate seizure control,” which reflects a market that GSK would seemingly have wanted to access through an authorized generic. Id. at 7.

Similarly, another exhibit that Defendants cite to only notes that “both patients and Neurologists are extremely resistant to a change of drug therapies – even to generic formulations of the same drug.” D.E. 479-2 at 11. The exhibit does not indicate GSK would not have pursued an authorized generic in addition to the Contracting Strategy. To the contrary, the document appears to indicate that there was an entirely new market to be serviced by the authorized generic, that is, for new patients or existing patients whose seizures were not sufficiently controlled. Defendants’ further contend that because customers participating in GSK’s Contracting Strategy “agreed to ‘dispense branded Lamictal® as a generic’ . . . a pharmacy could either sell

21 The Court assumes that “AEDs” refers to antiepileptic drugs. the [authorized generic] or participate in the Contracting Strategy, but not both, GSK would have illogically been competing against itself by launching an [authorized generic].” Opp. at 6 (emphasis in original). To support this proposition, Defendants cite to a GSK document, entitled “Post-generic Strategy for LAMICTAL®” that sets forth the following target: “Establish contract terms and conditions that will enable GSK to retain more of its market share by competing more

successfully following entry of generic lamotrigine.” D.E. 479-2 at 14. The document further reflects a proposal for “Retail and Mail Order” that provides “[a]llow discounts up to 40% off WAC to PBM22 mail order facilities and retail pharmacies from time of first generic entry to second generic entry.” Id. at 15. The document continues that “[m]ail order and retail pharmacies must demonstrate the capability and agree to dispense branded LAMICTAL® as a generic using a DAW-5 code or another similar mechanism for the same generic copay.” Id. The Court does not find Defendants’ argument convincing. As part of the Contracting Strategy, GSK required Lamictal to be dispensed as a generic. However, GSK was not legally required to do so. Instead, GSK was precluded from initially launching an authorized generic due

to the settlement agreement with Teva. Yet, this does not mean that if GSK did not face the same limitation (No-AG promise), it would have nevertheless engaged in the same course of conduct (dispensing Lamictal as a generic). In other words, absent the settlement agreement, GSK was free to launch its authorized generic and, as a result, free to change its Contracting Strategy so that it did not require Lamictal to be dispensed as a generic. In fact, in their initial briefing before Judge Walls, Defendants conceded that “GSK decided on the Contracting Strategy in 2008 – after the settlement already had taken an authorized generic off the table.” D.E. 406 at 30 (emphasis in original). When the Contracting Strategy required Lamictal to be dispensed as a generic, GSK

22 The Court assumes that “PBM” refers to a pharmacy benefits manager. could not compete with itself through an authorized generic because it was precluded from doing so under the settlement agreement. Defendants also rely on their expert witnesses’ reports. Br. at 6. Defendants’ expert Dr. Bruce E. Stangle opined that “it is unlikely GSK would have launched an authorized generic.” D.E. 479-2 at 22 ¶ 40. Dr. Stangle based this opinion first on the general proposition that “there

is a risk that the brand firm’s authorized generic will take sales away from the branded product, a phenomenon referred to as ‘cannibalization.’” Id. at 23-24 ¶ 41. Dr. Stangle reported that “cannibalization is an important consideration for GSK in evaluating whether to launch an authorized generic” and that the “brand firm is generally not entitled to all of the sales revenue from selling an authorized generic” but must share such revenue “with a marketing partner responsible for selling the product.” Id. However, Dr. Stangle did not cite any evidence as to Lamictal specifically. Dr. Stangle also reviewed evidence that GSK expected generic erosion to be lower and slower for Lamictal because of physicians’ concerns that users of the drug could suffer

“breakthrough seizures as a result of switching to a generic.” Id. at 23 ¶ 42. Dr. Stangle claimed that “these considerations, in particular expected higher brand retention rates [were] a strong disincentive to launch on authorized generic.” Id. at 24 ¶ 44. However, Dr. Stangle does not adequately explain his basis for this view. Unless GSK expected no significant generic erosion whatsoever, it appears that, as Plaintiffs’ experts explained, see D.E. 375-1 at 19 ¶ 24, GSK would still be incentivized to capture sales within the generic market if it had the right to do so. Indeed, as discussed above, GSK’s internal documents indicated that the Contracting Strategy was not focused on new patients or patients whose seizures were not well controlled. Dr. Stangle also pointed to other instances in which GSK did not launch an authorized generic out of concern for cannibalization. D.E. 479-2 at 25-27, ¶¶ 45-49. But as to Lamictal specifically, Dr. Stangle only relied on the testimony of Christopher Viehbacher,23 the former president of North American pharmaceuticals at GSK. Id. at 26 ¶ 48. Viehbacher testified that at the time of the No-AG Agreement, GSK had “no plans one way or the other,” that GSK “hadn’t

made any decision as to an authorized generic,” and that “there were a number of good reasons why you wouldn’t necessarily do an authorized generic with [Lamictal].” Id. This testimony does not establish that GSK had decided not to launch an authorized generic absent the No-AG Agreement; instead, the testimony reflects that GSK had not made a decision one way or the other. Last, Dr. Stangle pointed to the declaration of Bill Clark, “a Subject Matter Expert for Planning at GSK and former Supply Manager.” Id. at 27 ¶ 49. Dr. Stangle indicated that Mr. Clark had stated that “GSK struggled to maintain a supply of branded Lamictal for several reasons, including because GSK did not have, and could not get access to, enough . . . lamotrigine.” Id. Clark opined that because of the reduced quantities of lamotrigine and the fact that GSK would

have continued to manufacture brand Lamictal, he was “confident that GSK would have elected to continue manufacturing the branded product rather than making bailment stock of authorized generic.” Id. at 28 ¶ 51. However, Clark did not indicate when GSK was experiencing “the short supply” of lamotrigine. Moreover, GSK had been the exclusive supplier of the drug before Teva entered the market, and Defendants provide no evidence that GSK ever had difficulty in meeting

23 As with Dr. Lamb’s testimony, it appears that this portion of Dr. Stangle’s testimony is not within the purview of proper expert opinion. The same is true of Dr. Stangle’s opinion as to Bill Clark’s statements. Again, defense counsel is free to point to such evidence – that is, statements or testimony by GSK representatives – and argue their import. Yet, as the Court also did in Dr. Lamb’s case, the Court considers the underlying statements because they are part of the record. the supply demands of the entire market. Clark does not adequately explain how GSK would have been unable to supply fewer purchasers after lamotrigine was launched. Defendants also rely on the expert reports of Drs. Snyder and Hughes. Dr. Snyder opined that “the record does not support Professor Elhauge’s assumption that GSK was 100 percent certain to launch an AG” for nearly identical reasons as those expressed by Dr. Stangle: (1) generic

cannibalization; (2) GSK’s failure to launch authorized generics in other circumstances; and, (3) supply and manufacturing constraints. See D.E. 479-2 at 37-39 ¶ 33. However, Dr. Snyder also indicated that it would be plausible to estimate the probability of GSK launching an authorized generic at between zero percent to 50 percent. Id. at 39 ¶ 34. Dr. Hughes’ opinion that “[t]he evidence indicates that GSK would not have launched an authorized generic in the but-for world” is based on the same evidence relied on by Drs. Snyder and Stangle. See D.E. 479-2 at 41-45, ¶¶ 43-47. The Court finds that Plaintiffs have shown by a preponderance of the evidence that they can prove through common evidence that GSK, absent the settlement agreement, would and could

have pursued both the Contracting Strategy and an AG. Frankly, the Court does not see any apparent individual proof issues as to this evidence – either GSK would have done so or it would not have. Plaintiffs have expert testimony from a former pharmaceutical industry employee that (1) “GSK would have been ready, willing, and incentivized to simultaneously launch an authorized generic version of Lamictal tablets, and would have done so,” D.E. 374-4 at 6, ¶ 11; (2) “based on its previous AG launches, GSK had determined that it made financial sense to launch an AG and planned to do so,” id.; and, (3) launching an authorized generic and engaging in the Contracting Strategy are “complimentary” because “[a]n authorized generic allows the branded company to capture revenues from sales that would otherwise go to the generic company [and the] Contracting Strategy targets revenues from retained brand sales.” D.E. 375-1 at 19 ¶ 24. And GSK’s stated reason for implementing the Contracting Strategy – that physicians would be highly reluctant to switch patients who were stable on Lamictal – does not account for the new patient population to be served by the generic. GSK’s evidence makes clear that GSK expected “new patients or those who have yet to achieve adequate seizure control,” D.E. 479-2 at 7, to use a

generic and it is, at best, unclear why GSK would not have competed for that market through an authorized generic if it had the right to do so.24 C. Common Evidence as to Classwide Antitrust Injury The Court now addresses whether Plaintiffs have proven by a preponderance of the evidence that they can prove classwide antitrust injury through the use of common evidence. In doing so, the Court considers Plaintiffs use of averages in light of the answers to the three predicate questions set forth by the Circuit. The Court also reviews additional, related arguments raised on remand by the parties. As to the predicate questions, the Court has determined that neither side has proven by a

preponderance of the evidence that the market vis-à-vis lamotrigine was characterized by individual negotiations. Instead, Defendants have demonstrated that Teva set different lamotrigine prices for different categories of purchasers. Plaintiffs have shown by a preponderance of evidence that they can prove through common evidence that GSK would have pursued an authorized generic absent the No-AG promise. In addition, Teva has shown that it preemptively lowered the prices of lamotrigine for most customers in response to GSK’s Contracting Strategy.

24 GSK’s anticipated success under the Contracting Strategy did not pan out. As Plaintiffs’ expert indicated, Teva’s generic lamotrigine had captured nearly 75% of the market share for both the brand and the generic by July 2008. D.E. 373 at 54-55, ¶ 82-83. While this finding supports Defendants’ position, it does not provide a definitive answer to the Court’s inquiry. First, those Teva customers whose lamotrigine prices were not reduced in response to the Contracting Strategy could prove through common evidence that they suffered antitrust injury. Second, for those customers whose prices were lowered, the key question becomes whether the prices would have been lowered further if GSK had introduced an authorized generic?

Of course, a response to this question depends on how a jury answers the inquiry as to whether GSK would have launched an authorized generic absent the No-AG promise. If a jury finds that GSK would not have launched an authorized generic, then it appears that Defendants prevail without further deliberation. If a jury reaches the opposite conclusion, then whether Teva would have reduced its prices further – across the class – in the face of authorized generic competition becomes the critical issue. For purposes of the Court’s current analysis, as mandated by the Circuit, the important question is whether Plaintiffs can demonstrate through common evidence antitrust injury for each class member. 1. The Third Circuit’s Suboxone Decision

In their reply, Plaintiffs cite to the Third Circuit’s opinion in In re Suboxone Antitrust Litig., 967 F.3d 264, 272 (3d Cir. 2020). Reply at 2-3, 5. In Suboxone, the direct purchasers of Suboxone alleged that the manufacturer impeded generic entry in violation of § 2 of the Sherman Act. Suboxone, 967 F.3d at 267. Because its market exclusivity period for Suboxone tablets was set to expire, the manufacturer developed an under-the-tongue film of Suboxone. Id. at 267-68. The plaintiffs alleged the manufacturer attempted to eliminate the market for generic tablets by engaging in a widespread campaign to disparage the tablets’ safety, announcing that it was withdrawing Suboxone tablets from the market for safety reasons, ending its rebate program for Suboxone tablets in favor of one for the film, and manipulating the Food and Drug Administration’s review process through a baseless citizen petition. Id. at 268. The plaintiffs were all direct purchasers of Suboxone tablets. Id. The plaintiffs’ expert was also Dr. Lamb. Id. The district court certified the class. Id. at 269. On appeal, the manufacturer focused on Rule 23(b)(3)’s predominance requirement, first arguing that the plaintiffs did not suffer an antitrust injury because the manufacturer could lawfully

raise the price of Suboxone tablets and also change its rebate program. Id. at 270. The Third Circuit rejected this argument, reasoning that the plaintiffs’ theory of the case focused on the totality of the manufacturer’s conduct, not just the pricing changes and the rebate program. Id. The manufacturer next argued that predominance was not met because the plaintiffs’ model only calculated aggregate damages such that individual damages’ questions defeated predominance. Id. at 271. The Circuit again disagreed, stating the following: [The manufacturer] is incorrect. Antitrust plaintiffs may satisfy the predominance requirement by using a model that estimates the damages attributable to the antitrust injury, even if more individualized determinations are needed later to allocate damages among class members.

Id. (citing In re Modafinil Antitrust Litig., 837 F.3d 238, 262 (3d Cir. 2016), as amended (Sept. 29, 2016)). Suboxone is inapposite. The relevant issue before the Suboxone court was damages, not antitrust injury. If Plaintiffs are successful in showing that they can prove antitrust injury through common evidence, then Suboxone would be relevant as to Plaintiffs use of an aggregate damages model. 2. Parties Related Arguments on Remand The parties also raise additional, related arguments. At the outset, the Court addresses several of Plaintiffs’ assertions that appear to be made despite the Third Circuit’s remand opinion. First, Plaintiffs claim that there is common evidence to rebut Defendants’ Contracting Strategy defense. Br. at 4. But that is not the issue before the Court. Instead, the Court must decide whether Plaintiffs’ can prove antitrust injury to each class member through common evidence. Among the common evidence, Plaintiffs point to Teva’s forecasting documents and economic literature. Br. at 4-5. Yet, the Circuit cited to the same evidence and nevertheless remanded the matter. The

Circuit noted that Dr. Lamb relied on economic literature, Teva’s own general pricing forecasts, and transaction-level data reflecting the average price Teva’s customers paid. Lamictal, 957 F.3d at 193. Plaintiffs also rely on common evidence that prices for lamotrigine generally moved in unison, downward, upon the entrance of additional generics to the market. D.E. 373 at 50-52, ¶¶ 83-85. This is an accurate statement, but its relevance is diminished by the underlying facts. Here, there was no period when two generics competed on the market; instead, the number of generics increased from one to eight once Teva’s exclusivity ended. D.E. 406-3 at 10 ¶ 13; D.E. 373 at 22-23, ¶ 30; see also Br. at 5, n. 6 (“[T]he number of generics increased from one (Teva) to

eight.”)). Dr. Lamb, Plaintiffs’ expert, acknowledged that there is a cost savings increase “as the number of generics in the market increases until there [sic] [are] four or five generics in the market” and that “the price [will not go] materially lower with, for example, ten generics on the market as opposed to nine.” D.E. 373 at 41-42 ¶ 65(a); id. at 42, n. 1. Thus, when Plaintiffs argue that prices moved in unison with additional generics, they omit that the price drop was not gradual; rather, the market was flooded with eight generics at the same time which, as Plaintiffs’ expert admitted, caused the generic lamotrigine price to reach its nadir. This evidence does not provide any indication as to the lamotrigine price that would have existed for all class members in the but-for, two-generic world. Plaintiffs claim that common evidence shows that the price drop attributable to the Contracting Strategy “is much shallower than the generic price drop that would have occurred absent Defendants’ unlawful conduct.” Br. at 15-16. Plaintiffs rely on the “New Product Share/Pricing Assumptions – TUSA Generics East.” Br. at 16; see also D.E. 473 at 66. This pricing matrix shows that in year 1, if Teva was the exclusive generic, it planned to price that

generic at approximately 50% of the Lamictal’s WAC. Id. However, if Teva was competing with another generic (including an authorized generic), Teva projected its price at 35% of the brand’s WAC. Id. at 66 n. 198. The pricing matrix reflects Teva’s general forecasts over its entire generic portfolio and is not specific to lamotrigine. Defendants respond that the pricing matrix actually assumed a 44% WAC (because of retroactive discounts Teva would have to provide beginning in the seventh month) and that the matrix reflects a general rule of thumb as to all generic products. Opp. at 15, n.9. Defendants continue that the matrix does not reflect prices that Teva actually charged for lamotrigine. Opp. at 15. Nevertheless, to determine whether Teva followed the pricing matrix for lamotrigine, Dr. Lamb “compared the net price for generic lamotrigine tablets charged

by Teva during its six-month period of exclusivity to the brand WAC for Lamictal tablets during this time.” D.E. 473 at 67, ¶ 104. Dr. Lamb determined that “the weighted average ratio of ASP25 to brand WAC was 49.672 percent” which is consistent with the projected 50% discount reflected in Teva’s pricing matrix. Id. Plaintiffs therefore claim that, had GSK launched an authorized generic, the price of Teva’s generic would have been 30% lower than the price Teva originally projected it would charge – 35%, as opposed to 50%, of brand WAC. Plaintiffs further contend that, based on the analysis of

25 “ASP” refers to the “average selling price.” D.E. 373 at 65, ¶ 102. Dr. Hughes, Defendants’ expert, generic prices would have dropped even further “by at least 6.1%” because brand Lamictal’s WAC at the but-for generic entry date was 6.1% lower than in June 2008 – the month before Teva launched lamotrigine. Dr. Hughes’ report opined as follows: The brand WAC was 6.1 percent lower in September 2007 (the month before Teva would have launched its generic product in Dr. Lamb’s “No-Payment Settlement” scenario) than it was June 2008 (the month before Teva launched its generic product in the actual world). Following Dr. Lamb’s assumption that Teva set generic prices based on the brand WAC at the time of generic entry, generic prices would have been 6.1 percent lower in a but for world in which Teva launched in October 2007 than they actually were at the time of generic entry.

D.E. 478-3 at 5, ¶ 54. Using the common evidence of Teva’s forecasting documents – shown to be reliable based on a comparison to prices paid in the real world, see D.E. 473 at 67, ¶ 104 – and Dr. Hughes’ testimony, Plaintiffs assert that Plaintiffs would have paid, on average, 34.3%26 less for lamotrigine absent the No-AG Agreement. Br. at 18. Plaintiffs claim this price drop is larger “than the largest alleged preemptive Teva price drop (20.8%) than Defendants attribute to GSK’s [Contracting Strategy].” Id. at 18. Defendants respond that “[l]ike the failed sensitivity analysis, this theory compares a dizzying assembly of averages.” Opp. at 19. Defendants argue that “some purchasers would receive higher or lower discounts” and that “a given purchaser might have received a 20% discount off the generic price in the but-for world” and would therefore “be uninjured, not withstanding the higher average but-for discount.” Opp. at 19-20.

26 Plaintiffs actually assert that the discount is 36.1%. They arrive at this number by adding 30% (the discount projected by Teva) and 6.1% (the additional discount projected by Dr. Hughes). Br. at 17-18. But, as Defendants accurately point out, “percentages with different bases cannot be summed” so the appropriate figure is “34.3%.” Opp. at 19. Plaintiffs have a sound theory. Plaintiffs assert that if Teva had faced authorized generic competition when it launched lamotrigine, generic purchasers would have paid less. The economic literature supports this theory. While GSK indicates that pursuant to the Contracting Strategy, it lowered the price for Lamictal up to 40% for certain customers, the Court does not know which customers received which discount. In fact, other evidence strongly suggests that GSK did not

lower Lamictal’s price commensurate to that of an authorized generic. First, GSK believed that the unique nature of Lamictal meant that physicians would not (or would be reluctant to) switch patients to lamotrigine if the patients were stable on Lamictal. This unique feature would weigh against lowering the price of Lamictal to that of an authorized generic. Second, GSK appeared to internally acknowledge that it would lose a portion of the market to generic competition, particularly the market consisting of new patients or patients who were not yet stabilized. If GSK had lowered the price of Lamictal to that of an authorized generic, GSK should have been able to compete in this market. Finally, as noted, in November 2008, Teva’s President and CEO stated that the Contracting Strategy “was largely unsuccessful and really didn’t impact price all that

much.” D.E. 500-1 at 9. Plaintiffs’ theory, however rational it may be, is missing critical supporting evidence. First, while Plaintiffs rely on the evidence as to 50% of brand WAC without other generic competition compared to the 35% of WAC when facing a generic competitor, the Circuit was well aware of this evidence. Lamictal, 957 F.3d at 193 (indicating that Dr. Lamb relied on, among other things, “Teva’s own general pricing forecast tending to discount a generic by 50% without competition, but by 65% when facing an additional competitor”). Defendants respond (in accord with the Third Circuit) that its customers received varying discounts on lamotrigine in response to the Contracting Strategy, and that Plaintiffs have not shown that all such customers would have received an additional discount if GSK had launched an authorized generic. According to Defendants (and the Circuit), this is the crucial issue that the Court must confront in determining whether Plaintiffs’ can rely on averages in establishing classwide antitrust injury. And Plaintiffs have failed to prove that such evidence exists by a preponderance of the evidence. Plaintiffs have not provided the critical evidential link. For example, Plaintiffs did not

produce evidence showing that if Teva faced an authorized generic and the Contracting Strategy at the same time, it would have further lowered the prices of lamotrigine across the board. Nor did Plaintiffs, for instance, cite to evidence demonstrating that the Teva customers who received discounts in response to the Contracting Strategy, would have received additional, commensurate discounts in response to authorized generic competition. The Court assumes that Teva provided discounts for rationale business reasons rather than charitable, idiosyncratic, or arbitrary motivations. This assumption would mean that Teva would do the same in response to authorized generic competition. But this is an assumption, not evidence, and Plaintiffs have the burden of producing such evidence and proving the issue by a preponderance of the evidence.

As discussed, a key concern of the Third Circuit was the Court’s analysis of the competing experts vis-à-vis the use of averages. The Circuit noted, among other things, that Defendants’ expert, Dr. Hughes, criticized Plaintiffs’ expert’s, Dr. Lamb, use of general forecasting documents rather than lamotrigine-specific prices and indicated that Hughes developed his own model based on lamotrigine-specific prices from Teva documents that demonstrated that “25 of the 33 generic- only purchasers likely paid the same or lower prices in the actual world under the Contracting Strategy than they would have paid had GSK launched an AG.” Lamictal, 957 F.3d at 194. Dr. Hughes indicated that an “FTC study, cited by Dr. Lamb, shows that on average, the launch of an authorized generic decreases the generic’s product’s price by 6.6 to 13.5 percent. The FTC-reported discounts provide a reasonable benchmark for comparison to Teva’s price discounting in response to the Contracting Strategy[.]” D.E. 476-6 at 4, ¶ 52 (citing U.S. Federal Trade Commission, “Authorized Generic Drugs: Short-Term Effects and Long-Term Impact,” August 2011). Using this benchmark, Dr. Hughes then reached the following conclusions: Exhibit 4b demonstrates that the range of Contracting Strategy discounts received by generic-only purchasers in the proposed Class overlaps substantially with the range of authorized generic discounts reported by the FTC. In fact, Exhibit 4b shows that 15 generic-only purchasers actually received discounts larger than 13.5 percent, the higher of the two estimates reported by the FTC. An additional ten purchasers (i.e. 25 in total) received discounts larger than 6.6 percent, the lower FTC estimate.

Id. (footnotes omitted). Dr. Stangle, another defense expert, noted that average discount provided to Teva’s customers in response to the Contracting Strategy was 11%. D.E. 406-3 at 31 ¶ 54. Dr. Hughes calculated the following as to Teva’s customers in response to the Contracting Strategy: at least 4 did not receive a discount, 10 received discounts of 9.3%, 2 received discounts of 16.7%, and 13 received discounts of 20.8%.27 D.E. 406-2 at 79-80. Dr. Hughes’s conclusion – that “25 of the 33 generic-only purchasers likely paid the same or lower prices in the actual world under the Contracting Strategy than they would have paid had GSK launched an AG” – was emphasized by the Circuit. Lamictal, 957 F.3d at 194. On remand, Plaintiffs have not shown by a preponderance of the evidence that that they can prove through common evidence that all of Teva’s purchasers would have received additional

27 Plaintiffs argue that the Court should disregard Drs. Hughes and Stangle’s conclusions because their opinion are based on the Go Forward Pricing tab rather than Sheet 1 of TEVA00024341. Reply. at 8-10. Plaintiffs, however, do not provide additional analysis as to which specific conclusions by defense experts are invalid in light of Sheet 1. As a result, the Court declines to adopt Plaintiffs’ all-or-nothing approach. discounts had GSK also launched an authorized generic. Plaintiffs theory is reasonable, but they are missing the critical evidential link emphasized by the Circuit. I. CONCLUSION Plaintiffs’ motion for class certification as to generic-only purchasers is denied. An appropriate Order accompanies this Opinion. Dated: April 9, 2021 Qw VO®. Ni Da f John\Michael Vazquez, U.S.D/J.