The State of Connecticut; The State of Alaska; The State of Arizona; The State of Arkansas; The State of California; The State of Colorado; The State of Delaware; The District of Columbia; The State of Florida; The State of Georgia; The Territory of Guam; The State of Hawaii; The State of Idaho; The State of Illinois; The State of Indiana; The State of Iowa; The State of Kansas; The Commonwealth of Kentucky; The State of Louisiana; The State of Maine; The State of Maryland; The Commonwealth of Massachusetts; The State of Michigan; The State of Minnesota; The State of Mississippi; The State of Montana; The State of Nebraska; The State of Nevada; The State of New Hampshire; The State of New Jersey; The State of New Mexico; The State of New York; The State of North Carolina; The State of North Dakota; The Commonwealth of the Northern Mariana Islands v. Joseph R. Biden, Jr., President of the United States; Merrick B. Garland, Attorney General of the United States; Alejandro N. Mayorkas, Secretary of the Department of Homeland Security; and the United States of America

Opinion

UNITED STATES DISTRICT COURT DISTRICT OF CONNECTICUT

THE STATE OF CONNECTICUT; THE STATE OF ALASKA; THE STATE OF ARIZONA; THE STATE OF ARKANSAS; THE STATE OF No. 3:20-cv-00802-MPS CALIFORNIA; THE STATE OF COLORADO; THE STATE OF DELAWARE; THE DISTRICT OF

COLUMBIA; THE STATE OF FLORIDA; THE STATE OF GEORGIA; THE TERRITORY OF GUAM; THE STATE OF HAWAII; THE STATE OF IDAHO; THE STATE OF ILLINOIS; THE STATE OF INDIANA; THE STATE OF IOWA; THE STATE OF KANSAS; THE COMMONWEALTH OF KENTUCKY; THE STATE OF LOUISIANA; THE STATE OF MAINE; THE STATE OF MARYLAND; THE COMMONWEALTH OF MASSACHUSETTS; THE STATE OF MICHIGAN; THE STATE OF MINNESOTA; THE STATE OF MISSISSIPPI; THE STATE OF MONTANA; THE STATE OF NEBRASKA; THE STATE OF NEVADA; THE STATE OF NEW HAMPSHIRE; THE STATE OF NEW JERSEY; THE STATE OF NEW MEXICO; THE STATE OF NEW YORK; THE STATE OF NORTH CAROLINA; THE STATE OF NORTH DAKOTA; THE COMMONWEALTH OF THE NORTHERN MARIANA ISLAND; THE STATE OF OHIO; THE STATE OF OKLAHOMA; THE STATE OF OREGON; THE COMMONWEALTH OF PENNSYLVANIA; THE COMMONWEALTH OF PUERTO RICO; THE STATE OF RHODE ISLAND; THE STATE OF SOUTH CAROLINA; THE STATE OF TENNESSEE; THE STATE OF UTAH; THE STATE OF VERMONT; THE COMMONWEALTH OF VIRGINIA; THE STATE OF WASHINGTON; THE STATE OF WEST VIRGINIA; THE STATE OF WISCONSIN; and U.S. VIRGIN ISLANDS, Plaintiffs, v. SANDOZ, INC.; ACTAVIS HOLDCO US, INC.; ACTAVIS ELIZABETH LLC; ACTAVIS PHARMA, INC.; AMNEAL PHARMACEUTICALS, INC.; AMNEAL PHARMACEUTICALS, LLC; ARA APRAHAMIAN; AUROBINDO PHARMA U.S.A., INC.; BAUSCH HEALTH AMERICAS, INC.; BAUSCH HEALTH US, LLC; MITCHELL BLASHINSKY; DOUGLAS BOOTHE; FOUGERA PHARMACEUTICALS INC.; GLENMARK PHARMACEUTICALS INC., USA; JAMES (JIM) GRAUSO; GREENSTONE LLC; G&W LABORATORIES, INC.; WALTER KACZMAREK; ARMANDO KELLUM; LANNETT COMPANY, INC.; LUPIN PHARMACEUTICALS, INC.; MALLINCKRODT INC.; MALLINCKRODT LLC; MALLINCKRODT plc; MYLAN INC.; MYLAN PHARMACEUTICALS INC.; KURT ORLOFSKI; MICHAEL PERFETTO; PERRIGO NEW YORK, INC.; PFIZER INC.; SUN PHARMACEUTICAL INDUSTRIES, INC.; TARO PHARMACEUTICALS USA, INC.; TELIGENT, INC.; ERIKA VOGEL-BAYLOR; JOHN WESOLOWSKI; and WOCKHARDT USA LLC,

Defendants. RULING ON MOTION FOR SUMMARY JUDGMENT In this action, the Attorneys General of most of the States and several U.S. territories (“the States”) allege that thirty-six pharmaceutical companies and executives (“the Defendants”) each participated in a series of conspiracies, as well as an overarching conspiracy, to fix prices, allocate markets, and rig bids in the sale of dozens of generic drug products. Defendants Mylan Inc. and Mylan Pharmaceuticals Inc. (collectively, “Mylan”) have moved for summary judgment on all claims that they participated in any product-specific or overarching conspiracy. ECF No. 1001; see also ECF No. 1008 (Defendants’ Joint Memoranda in Support of Defendant-Specific Motions for Summary Judgment). For the reasons set forth below, I deny Mylan’s motion for summary judgment. I. BACKGROUND A. Procedural Background This is one of three cases in which the Attorneys General of the States have sued scores of defendants in the generic drug industry for alleged antitrust violations and unfair trade practices.

All three cases were originally filed in this Court but were transferred to the Eastern District of Pennsylvania (the “MDL Court”), which was designated by the Judicial Panel on Multidistrict Litigation (the “JPMDL”) to preside over these and other similar cases brought by private parties in a consolidated proceeding. ECF No. 9.1 In April 2024, these three cases were remanded to this Court by the JPMDL and assigned to me. ECF No. 11. The operative complaint in this case, ECF No. 196, alleges collusion in the pricing, market allocation, and bidding for some eighty generic drugs, chiefly for dermatological applications. The parties refer to it as “the Dermatology complaint.” In a previous ruling, I denied the Defendants’ Joint Motion for Summary Judgment on Overarching Conspiracy Claims, ECF No. 609, finding that there is evidence in the record—

chiefly testimony from cooperating witnesses—from which a reasonable jury could find that the alleged overarching conspiracy exists. Connecticut v. Sandoz, Inc., No. 20-cv-802, 2025 WL 3470502 (D. Conn. Dec. 3, 2025). I found that at least some of the Defendants had at least a tacit agreement that they would conduct business in accordance with the broadly understood industry “rules of engagement,” which included following any price increase by a rival firm and avoiding “poaching” each other’s customers. Id. at *17. The conspirators considered a competitor to be

1 Unless otherwise indicated, all ECF numbers in this ruling refer to entries on the docket of this case, not the same case when it was before the MDL Court, and each page number refers to the page number shown on the ECF stamp on the top of the cited page, not the page of the relevant brief or pleading designated by the parties. References to page numbers of documents that were filed under seal refer to the page number of the PDF file. “rational” or “reasonable” if it would “fall in line with those types of tactics.” Id. at *6. In concluding that there is a question of fact about whether the overarching conspiracy exists, however, I did not determine whether each corporate Defendant was a member of the overarching conspiracy. Shortly before I issued that ruling, most Defendants filed additional,

Defendant-specific summary judgment motions, many of which, including this one, contest whether a particular Defendant joined the overarching conspiracy.2 In addition to this motion, the Defendants have filed a Joint Memorandum that addresses the legal standard and argues that Defendants are entitled to summary judgment on particular state-law claims. ECF No. 1008. I will also consider that filing to the extent it is applicable to Mylan. B. Factual Background In the overarching conspiracy ruling, I discussed the facts underpinning the States’ claims against all corporate Defendants, and I incorporate that description here. See Sandoz, Inc., 2025 WL 3470502, at *2–14. I now address the facts specific to Mylan’s participation in the alleged conspiracies. The following facts are drawn from the parties’ Local Rule 56 Statements of Facts,

including the States’ Statement of Additional Material Facts (“SAMF”), ECF No. 1019-1, and the record. The facts are undisputed unless otherwise noted. Of the eighty drugs at issue in the complaint, the States allege that Mylan conspired with

2 As noted in the previous ruling, “[w]hether a particular Defendant was a member is a distinct issue from the issue whether a conspiracy existed at all.” Sandoz, Inc., 2025 WL 3470502, at *16 (citing 4 Sand et al., Modern Federal Jury Instructions–Civil ¶ 79-8 (“[Y]ou should first determine . . . whether the conspiracy existed. If you conclude that the conspiracy did exist, you should next determine whether each defendant was a knowing member of the conspiracy.”); id. ¶ 79-6 (“In order to prove the [price-fixing] conspiracy, . . . [t]he plaintiff does not have to show that . . . all of the persons alleged to have been members of the claimed conspiracy were in fact members.”)). The States do not seek to hold the individual defendants (for the most part employees and executives of the corporate Defendants) jointly and severally liable for their participation in the alleged overarching conspiracy. Id. at *1 n.1. its competitors to allocate customers, fix prices, and rig bids for three drugs: Pioglitazone HCL Metformin HCL tablets of both 500 mg and 850 mg (collectively, “Pio Met”), Bromocriptine Mesylate 2.5 mg tablets (“Bromocriptine”), and Phenytoin Sodium Extended Release 100 mg capsules (“Phenytoin”). ECF No. 1019-1 ¶ 2. The States also allege that Mylan is jointly and

severally liable for anticompetitive conduct related to all drugs at issue, including those it never sold or manufactured, because Mylan and the other corporate Defendants had an “overarching understanding to avoid competing with each other and to instead settle for what these competitors refer to as their ‘fair share’” of a given drug market. ECF No. 196 ¶ 5; see ECF No. 1019-1 ¶ 1. 1. Information Sharing Agreement As discussed in the recent ruling on the overarching conspiracy, Della Lubke, a director of national accounts at Sandoz and one of the States’ cooperating witnesses, testified that she had an arrangement with a Mylan employee to exchange competitively sensitive information during the period of alleged anticompetitive conduct. Sandoz, Inc., 2025 WL 3470502, at *12; SAMF ¶ 16 n.7. Her contact was James Nesta, Mylan’s Vice President of Sales and a member of the company’s Pricing Committee. ECF No. 1019-1 ¶ 11; SAMF ¶¶ 20, 31–32. Lubke testified that

every communication she had with Nesta was to “talk about competitive situations,” and that Nesta began “talking about competitive information immediately” on the “first call [she] had [with] him.” SAMF ¶ 31. Lubke said that she would pass along information she learned from Nesta to her superiors and colleagues working on pricing, including executives Armando Kellum and Michael Vezza, as the “point of all the information” was to inform Sandoz’s decisions by increasing revenue. The arrangement with Nesta was “very helpful to Sandoz internally,” Lubke testified. Id. ¶ 32. Lubke also testified that she and Nesta “understood how we were to use the information that we exchanged.” ECF No. 964-1 at 4440. Mylan, Lubke said, “didn’t like . . . other competitors coming after their customers. If you come after our customers, we will come after yours. [Nesta] said prices should be kept high. So there was definitely an understanding that I would use the information not to hurt Mylan.” Id. If Mylan ever used that information to harm Sandoz, Lubke said, she would have stopped sharing sensitive information. Id.

Lubke also testified that “the traits of a responsible company would be following price increases, not punishing a competitor for raising their prices by taking unfair market share, not tanking price just in order to get market share.” Id. Lubke said that her “impression is that Mylan was considered responsible” by Sandoz and other competitors in the generic pharmaceutical industry. Id. Lubke also testified, however, that she did not know what Nesta did with sensitive information that she shared with him. Id. She also said that she had no “personal knowledge of an overarching conspiracy” between Sandoz and all of the other Defendants “in which each company would get its fair share of particular generic markets.” Id. at 4258. When asked about an arrangement with Lubke to share competitive information, Nesta

invoked his right against self-incrimination under the Fifth Amendment. SAMF ¶ 36; ECF No. 964-1 at 5006–19, 5021–26. Sandoz executive Kellum, a Defendant in this case, also invoked the Fifth Amendment when asked about information obtained from Mylan, including in response to a question about whether he “made sure that Sandoz followed through on Ms. Lubke’s commitments to Mr. Nesta.” ECF No. 964-1 at 3212. 2. Pio Met Pio Met is used to control high blood sugar in patients with type 2 diabetes. ECF No. 1019 ¶ 3. In early 2013, Mylan, the first-to-file generic manufacturer, was nearing the end of a 180-day exclusivity period in which its only other competitor was non-Defendant Teva, which was authorized by the brand-name supplier to sell a generic version of Pio Met. ECF No. 1019-1 ¶ 84. In February 2013, Mylan supplied fifty-two percent of the generic Pio Met market, with Teva supplying the other forty-eight percent. SAMF ¶ 1. After the exclusivity period ended, Aurobindo and another non-Defendant, Torrent, entered the market, and Sandoz followed in April 2013. ECF No. 1019-1 ¶ 84. During the alleged conduct period, Mylan and Teva lost share to the new entrants.

By the time of the filing of the Heritage complaint in late 2016, Aurobindo had the highest market share at just under forty percent, followed by Mylan at about thirty-five percent, Teva at about eighteen percent, and Torrent at about fifteen percent, while Sandoz had left the market. See ECF No. 959-1 at 679–80; ECF No. 968-2 at 34 fig.13. With the new entrants in the market, the average net price of Pio Met went up briefly in the third quarter of 2013 but then began to drop, and in 2016 was lower than when the alleged conduct began in 2013. ECF No. 959-1 at 679–80. In January 2013, Mylan analyzed its customers to determine which it planned to retain when new competitors entered following the exclusivity period. ECF No. 1019-1 ¶ 85. From January 29 to January 30, 2013, Mylan’s Nesta exchanged twenty-six calls with three potential competitors: Anthony Thomassey at Aurobindo, Kevin Green at Teva, and Tim Gustafson at

Ranbaxy. SAMF ¶ 2; ECF No. 946-1 at 20–21. He reported to colleagues in an email on January 29 that he “hear[d]” that “Aurobindo [is] coming.” SAMF ¶ 3. The next day, at 8:55 a.m., Nesta wrote in an email that Ranbaxy would not enter the market, and that Aurobindo was “looking for 20 [percent] share.” Id. He also wrote that information about others entering the market “will be critical information so we can maintain 3 player pricing.” Id.; ECF No. 959-1 at 188. One minute later, Nesta called Thomassey at Aurobindo, and the two exchanged more calls throughout the day. Id. ¶ 4; ECF No. 946-1 at 20–21; ECF No. 964-1 at 5149. In the final week of Mylan’s exclusivity window in February 2013, Nesta exchanged more calls with Aurobindo’s Thomassey and Teva’s Green. SAMF ¶ 5. After one call with Nesta on February 8, 2013, Thomassey immediately called his boss, James (Jim) Grauso, who was then Aurobindo’s Vice President of Sales and Marketing. Id. ¶ 6 & n.6. While on another call with Grauso an hour later, Thomassey received an email from Grauso that included the monthly Pio Met usage of a customer, Cardinal Health. Id. At that time, Cardinal was purchasing Pio Met

from Mylan at a contract price of $229.94. ECF No. 1019-1 ¶ 90. Thomassey then exchanged more back-to-back calls with Nesta and Grauso that afternoon. SAMF ¶ 6. While on a call with Grauso, Thomassey emailed Cardinal an offer to supply Pio Met. Id. On February 10, 2013, Nesta wrote in an email to a colleague that “[w]e should be good to go on [] Pio Met” because he had “locked all this up last week.” Id. ¶ 7. On February 11, 2013, Thomassey spoke again with Nesta and then immediately emailed Grauso. Id. ¶ 8. In the email, Thomassey wrote that he had spoken with a Cardinal Health employee, who purportedly “said it will be a couple of days before they have an answer. The right of first refusal has been sent off to Mylan.” ECF No. 951-1 at 71. The next day, Aurobindo contracted with Cardinal Health to be its primary supplier of Pio Met at a contract price of $143.95. SAMF ¶ 9; ECF No. 1019-1 ¶ 90. On February 18, 2013, Mylan agreed

to serve as Cardinal’s secondary (i.e., backup) supplier of Pio Met, in the event Aurobindo could not supply, at a higher price. ECF No. 1019-1 ¶ 93. Thomassey provided testimony as a cooperating witness.3 During his deposition, Thomassey testified that “definitely it was” likely that he was talking to Nesta about Pio Met on their calls on February 8, 2013, as Aurobindo was preparing to enter the market, and that “they absolutely were not social calls.” ECF No. 965-1 at 459. Thomassey said that he was speaking with Nesta at Grauso’s direction “[t]o make sure that we split the market share appropriately and

3 As I noted in the overarching conspiracy ruling, the deposition transcripts of Thomassey in the summary judgment record raise no concerns about his memory or his competence as a witness, despite the Defendants’ objections. Sandoz, Inc., 2025 WL 3470502, at *6 n.6. keep the . . . price as high as possible.” SAMF ¶ 12; ECF No. 965-1 at 459–60. Thomassey also testified that, given the pattern of phone calls on February 8, it was likely that he had been speaking to Nesta about Aurobindo’s bid for Cardinal Health, which was then a Mylan customer. SAMF ¶ 12; ECF No. 965-1 at 460. Asked about the topic of discussion with Nesta on February 11,

Thomassey testified, “[i]t would have been about whether they’re going to give up Cardinal or not.” ECF No. 965-1 at 461. Thomassey then said there was an agreement between Mylan and Aurobindo “[t]hat [Mylan] would walk and [Aurobindo] would . . . get the award at Cardinal.” Id. at 462. When asked about these calls with Thomassey, Nesta invoked his Fifth Amendment right against self-incrimination. SAMF ¶ 22. Teva’s Green testified that his superior would often ask him to gather competitors’ information from his “buddy at Mylan,” Nesta. Id. ¶ 20. Green testified that Nesta would tell him “[w]hether [Mylan was] in a position to take market share, . . . whether [Mylan] felt they would follow if it was a price increase conversation, or if it was another market event like Teva launching, you know, would they – be holding on to a lot of their share.” Id. ¶ 21. Between January 28 and

February 13, 2013, Aurobindo’s Grauso also exchanged calls with Teva employees, including Green. Id. ¶ 14. On February 12, 2013, the same day that Aurobindo secured Cardinal as a customer, it also secured the business of another customer, Harvard, at a contract price of $131.95. ECF No. 1019- 1 ¶ 95. From February 7 to February 13, 2013, Aurobindo’s Grauso exchanged frequent calls with Teva’s Green and another Teva employee. SAMF ¶ 14; ECF No. 949-1 at 135–140, 148. On February 14, 2013, Teva offered Harvard a lower price of $122, which Aurobindo matched to retain the customer. ECF No. 1019-1 ¶ 95. The following day, February 15, 2013, Torrent offered Harvard a contract price of $90 to supply Pio Met, which Aurobindo again met to retain the customer. Id. ¶ 97. On the same day, Cardinal requested “a cost adjustment” from Aurobindo, also seeking a $90 price. Id. ¶ 98. Thomassey then emailed Grauso, writing in reference to Torrent, “I suggest we walk….. maybe the morons will stop.” ECF No. 968-2 at 937. In his deposition, Thomassey testified that “a nickname for Torrent that they had in the industry was

‘torch it,’ because they were so bad at eroding prices. That's why I said, ‘maybe the morons will stop.’” ECF No. 965-1 at 575. Aurobindo offered a reduced price of $115.30, which Cardinal accepted. ECF No. 1019-1 ¶ 99. Neither Teva nor Torrent is a Defendant in this action. While Teva is a Defendant in one of the other generic pharmaceutical cases currently before me, No. 19-cv-710, the States have not alleged any wrongdoing by Torrent in this or any other action. Id. ¶ 97. On February 19, 2013, Thomassey spoke three times with a counterpart at Sandoz, Christopher Bihari, about Pio Met, according to phone records and contemporaneous notes taken by Bihari, who is another of the States’ cooperating witnesses. SAMF ¶ 15. Bihari testified that he and Thomassey spoke about Torrent driving down prices and Aurobindo being “at Cardinal

currently.” Id. Thomassey and Bihari continued to communicate into April, when they exchanged non-public contract pricing. Id.; ECF No. 964-1 at 919–20. Bihari testified that this exchange of information “allowed Sandoz to appropriately price [Pio Met] in the market upon launch and to target these customers.” ECF No. 964-1 at 920. Bihari also testified that Sandoz and Aurobindo had an agreement on Pio Met, in which Sandoz executives Kellum and Vezza participated, to “ensure Sandoz keeps their pricing high in the market upon entry and to target certain customers.” Id.; SAMF ¶ 19. But Bihari also testified that he did not enter into any agreements with anyone at Mylan concerning any products. ECF No. 1019-1 at 31. Kellum invoked his Fifth Amendment right against self-incrimination when asked whether Pio Met competitors, including Mylan, had an agreement to fix prices and allocate customers, and whether Sandoz and Aurobindo had a broader agreement not to compete on pricing in generic drug markets where they overlapped. SAMF ¶ 22; ECF No. 964-1 at 3386. On May 13, 2013, Mylan’s Nesta phoned Lubke at Sandoz. SAMF ¶ 16. The next day,

grocery chain Publix, a Mylan customer, informed a Mylan national account manager, Steve Krinke, that Sandoz had made a bid on Pio Met. Id. ¶ 17. Mylan’s Pricing and Contracts team (“P and C”) determined internally that Mylan would not reduce its price. Id. On May 20, 2013, Krinke emailed Nesta and noted that “P and C want to let this product go at Publix…I do not…help?” Id. Nesta responded, “Have to let go. New entrant. We have 40 share.” Id. ¶ 18. Later that day, Nesta phoned Lubke at Sandoz in a call that lasted six seconds. Id. This was one of ten calls between Nesta and Lubke between May 8 and May 31, 2013.4 Id. ¶ 16. 3. Bromocriptine Bromocriptine is used to treat Parkinson’s disease, hyperprolactinemia, and acromegaly. ECF No. 1019 ¶ 3. According to data analyzed by the parties’ experts, Mylan and Sandoz each had a roughly equal share of the market until Perrigo entered the market in 2009, at which point

Mylan’s share steadily decreased as Perrigo gained share. ECF No. 968-2 at 38 fig.15. When the alleged conduct began on April 5, 2013, Sandoz had a sixty-five percent market share, Perrigo had a thirty-percent share, and Mylan had a five percent share. SAMF ¶ 24 (noting this split in November 2012); see also ECF No. 965-1 at 1067 (same split in March 2013). In February 2013, Mylan planned to temporarily discontinue its Bromocriptine sales during a shortage of raw materials. ECF No. 1019-1 ¶¶ 12, 15. In March 2013, Mylan exited the market for two months because of its supply shortage. SAMF ¶ 24. On March 1, 2013, Walgreens, a

4 The States appear to assert that at least some of these calls between Nesta and Lubke involved discussion about Bromocriptine, rather than Pio Met. See SAMF ¶ 25. Mylan customer, asked Sandoz if it could temporarily supply Bromocriptine while Mylan was out of the market. ECF No. 1019-1 ¶ 16. Between March 4 and March 11, 2013, Mylan’s Nesta and Sandoz’s Lubke exchanged three telephone calls. SAMF ¶ 25. Although she had no specific recollection of the calls, when

shown relevant documents, Lubke testified it was “probable and consistent” with her communication practices with Nesta that these discussions pertained to the Bromocriptine supply issue and resulting market pricing. Id. ¶ 31; ECF No. 1019-1 ¶ 33. She acknowledged, however, that Sandoz already knew of Mylan’s supply issue before her calls with Nesta. ECF No. 1019-1 ¶ 33. On March 13, 2013, Mylan notified its customers that it was discontinuing Bromocriptine sales, and Sandoz received additional requests from customers on the same day. Id. ¶¶ 17–18. On March 22, 2013, an internal pricing committee at Sandoz recommended increasing the company’s price on Bromocriptine. Id. ¶ 20. On April 2, 2013, Sandoz’s Bihari called Anthony Polman of Perrigo because “[he] was informing [Polman] of [Sandoz’s] upcoming bromocriptine price increase.” SAMF ¶ 26. Immediately afterwards, Bihari called Kellum, his superior at

Sandoz, “[t]o ensure that [Kellum was] aware of the percent or price point and for Perrigo to follow.” Id. ¶ 27. The next day, Sandoz announced a 206 percent price increase, effective April 5, 2013. ECF No. 1019-1 ¶¶ 20–21; SAMF ¶ 29. Bihari said it was unusual for Sandoz to lead a price increase, but in this instance, he testified, “I believe it’s because Perrigo was the other competitor in the market, and they had a high level of confidence that Perrigo would follow and not take their customers.” ECF No. 964-1 at 922. Vezza, another cooperating witness and a Sandoz executive, testified that Sandoz “preferred to follow price increases, because it was easier to tell the pricing committee that we’re following. But, I mean, we did lead here and there.” ECF No. 965-1 at 1067. Vezza also said that there was less risk in leading a price increase when Perrigo or Taro was a competitor because Bihari “was communicating [with Perrigo and Taro] on a regular basis[, a]nd the goal or the agreement was to just follow the price increase. . . . [I]t was knowing that they took the price increase and we would follow as quickly as possible.” Id. at 1067–68.

As Mylan was preparing to resolve its supply issues, Nesta and Lubke exchanged ten phone calls between May 8 and May 31, 2013.5 SAMF ¶¶ 16, 25. On May 23, 2013, once Mylan had sufficient supply to reenter the Bromocriptine market, Mylan’s internal pricing team recommended relaunching Bromocriptine at Sandoz’s price. ECF No. 1019-1 ¶ 22.6 On the same day, Nesta approved this price increase and spoke twice with Lubke. SAMF ¶¶ 25, 30. On May 31, 2013, the day that the price increase was to go into effect, Nesta and Lubke spoke four times. Id. ¶ 25. Upon reentering the market, Mylan “recaptured” many of its previous Bromocriptine customers at the higher price. ECF No. 1019-1 ¶ 23. Perrigo was also experiencing supply issues in May and June 2013, however, and it did not promptly increase its Bromocriptine price after Sandoz’s increase. ECF No. 1019-1 ¶¶ 25–26.

Over a year later, in July 2014, Sandoz’s largest Bromocriptine customer, McKesson, asked Sandoz to match Perrigo’s price. ECF No. 965-1 at 1069–70. Internally, Vezza initially proposed adjusting Sandoz’s price, but just hours later, Sandoz informed McKesson that its price would not change. Id. at 1070. Asked why Sandoz did not adjust its price, Vessa testified: [S]omething must have happened. . . . [S]omewhere in the five-hour window . . . we, you know, were confident that we would not lose the award in some way so we bid the same market price. The only thing I could think of is there was some communication there where we, via [Bihari] . . . were given some sort of feedback

5 The States suggest that Nesta and Lubke were also discussing a different drug, Pio Met, on at least some calls during May 2013. See ECF No. 1019 at 9; SAMF ¶ 18. 6 In the complaint, the States alleged that Kellum called a contact at Mylan on May 23, 2013. The States later conceded, however, that they “do not have records of telephone calls between Armando Kellum and [said contact] on May 23, 2013.” ECF No. 1019-1 ¶ 28. that [Perrigo was] not going to go after McKesson. Otherwise, . . . I think we would more than likely try to reduce our price. Id. at 1070–71; see also id. at 1206 ( “I don’t know any other reason why I would have gone from let’s drop our price . . . to then not changing our price”). According to Bihari’s phone records from that day, he first spoke with Vezza, then called Perrigo’s Polman, after which he immediately called Vezza again. Id. at 1072; ECF No. 964-1 at 923. Bihari confirmed that when he spoke to Vezza he “relay[ed] . . . the information that Tony Polman had shared with me[;] they still weren’t . . . going to touch any of Sandoz’s customers. They were going to leave them alone.” SAMF ¶ 33. Perrigo’s Polman called Mylan’s Nesta on September 25, 2014. Id. ¶ 34. In his deposition,

Nesta invoked the Fifth Amendment in responses to questions about this call, as well as questions about any agreement between Mylan, Perrigo and Sandoz related to Bromocriptine.7 ECF No. 964-1 at 5020–21. On October 8, 2014, about two weeks after Polman and Nesta’s call and over eighteen months after Sandoz’s increase, Perrigo informed its customers of a Bromocriptine price increase. SAMF ¶ 34; ECF No. 959-1 at 388. Bihari called Polman on the same day and later sent an email to his team noting that “Perrigo increased WAC [price] to match ours and contract pricing is in line with the market.” SAMF ¶ 34. The States do not dispute, however, that Perrigo’s price was “approximately 33–40% lower” than that of Sandoz and Mylan. ECF No. 1019-1 ¶ 27; ECF No. 965-1 at 902. Still, Bihari said Sandoz “absolutely” had an agreement with Perrigo that Perrigo

would “follow Sandoz’s price increase, to ensure that their pricing [was] at parity with Sandoz, and not to touch any of Sandoz’s market share even though it took well over . . . a year for them to follow.” SAMF ¶ 35. Vezza testified that Perrigo likely “was doing whatever it is they had to

7 Polman died in 2018, before this action was filed. ECF No. 617-1 at 20. do to follow our increase” because “they were basically . . . leaving opportunity on the table for that entire time by not following us.” ECF No. 965-1 at 1073–74. Bihari and Vezza both testified that Sandoz was able to keep its Bromocriptine price higher because of Bihari’s communications with Perrigo. ECF No. 964-1 at 925; ECF No. 965-1 at 1072.

The economic data analyzed by the parties’ experts indicate that Mylan had an average quarterly net loss of $97,034 on Bromocriptine before the alleged conduct period, but a quarterly net profit of $212,862 during the conduct period, and a net profit of $174,318 after the conduct period. ECF No. 968-2 at 39 fig. 16. Mylan had a market share of about ten percent in the final quarter of 2016, before the alleged conduct period ended in December 2016, while Perrigo and Sandoz had a rough split of the remaining market. Id. at 38 fig. 15. Mylan’s expert, Dr. Lauren Stiroh, testified that these outcomes were “consistent with what we would expect to see under oligopolistic competition where some customers are yielded and some customers are maintained as more and more entrants come into a market environment.” ECF No. 964-1 at 6896; see also ECF No. 968-2 at 28–32 (Stiroh’s expert report). The States’s expert, Dr. Warren-Boulton, opined

that “the primary indicator of collusion is a large increase in price that is not explained by a change in market or structural conditions,” ECF No. 952-1 at 189. His report shows that the net price for Bromocriptine peaked at the start of the conduct period. ECF No. 959-1 at 629. He opined that Mylan saw overcharges—the difference between actual and but-for prices, calculated relative to average pre-conspiracy prices—and an increase in net revenues for Bromocriptine of at least $4.8 million during the conduct period. SAMF ¶ 61; ECF No. 952-1 at 151 n.163. 4. Phenytoin Phenytoin is used to prevent and treat seizures. ECF No. 1019 ¶ 3. According to data analyzed by the parties’ experts, in the final quarter of 2013, Taro had close to fifty percent of the market for generic Phenytoin, while Amneal had about thirty percent, Mylan fifteen percent, and Sun five percent. See ECF No. 959-1 at 697 fig.102; ECF No. 968-2 at 37 fig.14. Pfizer, the brand-name (i.e., not generic) manufacturer of Phenytoin also continued to sell the product. ECF No. 1019-1 ¶ 82. Amneal, Mylan, and Sun all had supply disruptions in 2014, leading to production delays and backorders. Id. ¶¶ 47–48. In March 2014, Amneal, which had low profit

margins on Phenytoin and was facing significant penalties for failing to supply its contract customers, informed Walgreens that it would likely have to raise prices or discontinue the product. Id. ¶ 50. Phone records show that the competitors communicated with each other during this period of shortage as Walgreens searched for a supplier. Taro executive Ara Aprahamian exchanged two phone calls with a Mylan sales account director, Michael Aigner, on April 10, 2014. SAMF ¶ 39. On April 16 and 22, 2014, Walgreens asked Aigner if Mylan could supply Phenytoin. ECF No. 1019-1 ¶ 51. Amneal then informed Walgreens on April 24, 2014, that it was increasing its Phenytoin price, effective June 23, 2014. Id. ¶ 52. From April 26 to April 29, 2014, Taro’s Aprahamian, Amneal’s Senior Director of Pricing Shannon Rivero, and Mylan’s Nesta, among

other company representatives, all attended the National Association of Chain Drug Stores Annual Meeting. SAMF ¶ 38. On April 30, 2014, an internal Mylan document indicated a “Potential Amneal price increase” for Phenytoin. Id. ¶ 40. On May 1, 2014, an Amneal vice president emailed a colleague that, in light of their difficulties producing Phenytoin, the company was “already working to [lose] Walgreens or Walmart” as a Phenytoin customer, with “plans . . . to keep one at a[n] improved price.” ECF No. 1019-1 ¶ 53; ECF No. 968-2 at 621. On May 8, 2014, Walgreens informed Mylan that it was increasing its estimated Phenytoin purchases by five times the previously requested amount. ECF No. 1019-1 ¶ 55. On May 14, 2014, in discussing the possibility of meeting this new demand, one Mylan analyst wrote, “I know we want the business,” before determining that Mylan could supply Walgreens starting July 21, 2014. Id. ¶ 56; ECF No. 968-2 at 649. On May 29, 2014, Mylan’s Aigner exchanged seven phone calls and one text message with Amneal’s Rivero. SAMF ¶ 47. The next day, Aigner sent Walgreens an offer that noted the July supply timeline. ECF No. 1019-1 ¶¶ 59–60. By June 10,

Walgreens had declined Mylan’s offer and remained Amneal’s customer at its higher price point. Id. ¶ 61; ECF No. 968-2 at 964. Pfizer and Taro increased their Phenytoin prices on June 1 and 2, 2014, respectively, with Taro doubling its price, going above even Amneal’s planned price increase to Walgreens. ECF No. 1019-1 ¶ 65–66; SAMF ¶¶ 48, 51. Citing published information about Taro’s increase, Mylan’s pricing team noted internally on June 3, 2014 that Taro’s new prices were “significantly above Mylan’s.” ECF No. 1019-1 ¶ 67. Upon learning of Taro’s increase on June 2, Amneal’s senior director Rivero emailed an Amneal director of pricing, Allen Lowther, writing, “Hope Mylan sees this.” SAMF ¶ 45. Lowther replied, “They will follow.” Id. Lowther, who had previously worked for Mylan, testified

that this statement was based on his understanding of Mylan’s “general business practice.” Id. ¶ 46. A flurry of calls between Amneal and Mylan followed on June 2: Amneal’s Rivero called Mylan’s Aigner in a call that lasted two seconds; Rivero then immediately called Mylan’s Nesta for four seconds, with more brief calls between the two following over the next hour; Amneal’s Lowther called Nesta later that evening, and the two exchanged more calls the following day. Id. ¶¶ 42, 44. Rivero and Aigner exchanged two more calls on June 4 and a brief call on June 5. Id. ¶ 48; ECF No. 945-1 at 80. On June 4, 2014, Amneal’s Lowther sent his colleague Rivero a document titled “Phenytoin Price Increase” that “[m]atched the new Taro” prices. ECF No. 968- 2 at 722. Between June 4 and June 9, 2014, Mylan’s Aigner spoke four times with Taro’s Aprahamian. SAMF ¶ 48; ECF No. 945-1 at 80. On June 18, 2014, Mylan’s team proposed increasing its own Phenytoin prices, citing Pfizer and Taro’s increases, and Nesta then approved

Mylan’s increase. ECF No. 1019-1 ¶¶ 69–70. Between June 3 and June 20, Nesta spoke on the phone nineteen times with either Rivero or Lowther from Amneal. SAMF ¶ 44. On June 25, 2014, Amneal’s Lowther sent an internal email that said, “We are matching Taro’s new [prices] and expect Mylan to do the same.” ECF No. 964-1 at 6269. Rivero testified that Lowther was “making an assumption that Mylan is going to change as well,” based on his experience working at Mylan. Id. at 6270. Amneal informed customers that it was following Taro’s price increase on July 1, 2014. SAMF ¶ 51. Amneal’s Rivero spoke with Mylan’s Aigner on that day and twice more before July 10. Id. ¶ 50. Aigner also spoke four times with Taro’s Aprahamian in July before Mylan followed Taro’s Phenytoin price increase on July 16, 2014. Id. ¶ 53. An employee of Sun also spoke with

Amneal’s Rivero in mid-June, before Sun increased its Phenytoin price on July 14, 2014. Id. ¶¶ 52-53. On July 9, 2014, Walmart requested that Mylan submit a bid to supply Phenytoin. ECF No. 1019-1 ¶ 72. Mylan’s pricing team conducted an analysis and, noting the product was on back order, concluded that Mylan could supply on September 1. Id. ¶¶ 73–74. On July 19, 2014, Mylan noted this timeline in submitting a bid that was roughly thirty-five percent less than Walmart was currently paying; Walmart declined the offer. Id. ¶ 75. Aigner, Nesta, and Aprahamian each invoked their right against self-incrimination under the Fifth Amendment when asked about Phenytoin or communications with competitors. SAMF ¶¶ 55–57. Rivero testified that she did not recall anything about calls during this time period, though she said that “others at Amneal were talking to employees of competitors.” Id. at ¶ 54. The economic data analyzed by the parties’ experts indicate that Mylan’s average quarterly profits from Phenytoin declined by $372,722 during the alleged conduct period, between Taro’s

increase on June 2, 2014, through the filing of the States’ Heritage complaint in December 2016. ECF No. 968-2 at 39. Dr. Warren-Boulton, however, opined that under one model, Mylan still saw overcharges and an increase in net revenues for Phenytoin during the conduct period relative to average prices pre-conspiracy. ECF No. 952-1 at 280. During the conduct period, Taro’s market share dropped from nearly fifty percent to about thirty-five percent, while Amneal gained market share. ECF No. 968-2 at 37. The market shares of Mylan, Sun, and Pfizer remained relatively stable during the conduct period. Id.; ECF No. 959- 1 at 193 & n.275, 697; SAMF ¶ 60. Dr. Stiroh opined that, if the competitors had an agreement to maintain a “fair share” of the market, “one would not have . . . expected to see the observed swings in sales share with Amneal, in particular, appearing to gain share during the purported Conduct

Period at the expense of Taro.” ECF No. 968-2 at 35. The report of Dr. Warren-Boulton, who again focused on a “large increase in price” as “the primary indicator of collusion,” ECF No. 952- 1 at 189, showed that the net price for Phenytoin began to rise at the start of the conduct period to a peak in early 2015. ECF No. 959-1 at 697. II. LEGAL STANDARD A. Summary Judgment “Summary judgment is appropriate only if the movant shows that there is no genuine issue as to any material fact and the movant is entitled to judgment as a matter of law.” Tolan v. Cotton, 572 U.S. 650, 656–57 (2014) (internal quotation marks and citations omitted). In reviewing the summary judgment record, a court must “construe the facts in the light most favorable to the non- moving party and must resolve all ambiguities and draw all reasonable inferences against the movant.” Caronia v. Philip Morris USA, Inc., 715 F.3d 417, 427 (2d Cir. 2013). “A genuine dispute of material fact exists for summary judgment purposes where the evidence, viewed in the

light most favorable to the nonmoving party, is such that a reasonable jury could decide in that party’s favor.” Zann Kwan v. Andalex Grp. LLC, 737 F.3d 834, 843 (2d Cir. 2013). The moving party bears the burden of demonstrating that no genuine issue exists as to any material fact. Celotex Corp. v. Catrett, 477 U.S. 317, 323–25 (1986). If the moving party carries its burden, “the opposing party must come forward with specific evidence demonstrating the existence of a genuine dispute of material fact.” Brown v. Eli Lilly & Co., 654 F.3d 347, 358 (2d Cir. 2011). “Where no rational finder of fact could find in favor of the nonmoving party because the evidence to support its case is so slight, summary judgment must be granted.” Id. (internal quotation marks omitted). B. Sherman Act The Sherman Act bans “[e]very contract, combination in the form of trust or otherwise, or

conspiracy, in restraint of trade or commerce among the several States.” 15 U.S.C. § 1. “The crucial question in a Section 1 case is therefore whether the challenged conduct stems from independent decision or from an agreement, tacit or express.” Starr v. Sony BMG Music Entm't, 592 F.3d 314, 321 (2d Cir. 2010) (internal quotation marks omitted). “An agreement between competitors to fix prices, known as a horizontal price-fixing agreement, categorically constitutes an unreasonable restraint, and, accordingly, is unlawful per se.” In re Publ'n Paper Antitrust Litig., 690 F.3d 51, 61 (2d Cir. 2012). At the summary judgment stage, “to raise a genuine issue of material fact as to an antitrust conspiracy, the plaintiff must present direct or circumstantial evidence that ‘tends to exclude the possibility that the alleged conspirators acted independently.’” Anderson News, L.L.C. v. Am. Media, Inc., 899 F.3d 87, 98 (2d Cir. 2018) (quoting Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 588 (1986)). “The quality of evidence required to satisfy the ‘tends to exclude’ standard ‘varies with the economic ‘plausibility’ of the alleged agreement.” In re Foreign

Exch. Benchmark Rates Antitrust Litig., No. 13 CIV. 7789 (LGS), 2022 WL 294118, at *5 (S.D.N.Y. Feb. 1, 2022). “[T]he tends to exclude standard is more easily satisfied when[] the conspiracy is economically sensible for the alleged conspirators to undertake and the challenged activities could not reasonably be perceived as procompetitive.” Anderson News, 899 F.3d at 99 (quoting Publ'n Paper, 690 F.3d at 63). A plaintiff need not “disprove all nonconspiratorial explanations for the defendants’ conduct; rather the evidence need be sufficient only to allow a reasonable fact finder to infer that the conspiratorial explanation is more likely than not.” Id. (internal quotation marks omitted). “[I]f the evidence is in equipoise, then summary judgment must be granted against the plaintiff.” Id. As noted in the ruling regarding the overarching conspiracy, Sandoz, Inc., 2025 WL

3470502, at *3, the parties agree that the suppliers in the markets for all the Drugs at Issue, including the three involved in this motion, comprise oligopolies. “Oligopolies pose a special problem under § 1 [of the Sherman Act] because rational independent actions taken by oligopolists can be nearly indistinguishable from horizontal price fixing,” as “any rational decision [in an oligopoly] must take into account the anticipated reaction of the other firms.” In re Mylan N.V. Secs. Litig., 666 F. Supp. 3d 266, 318 (S.D.N.Y. 2023). “Consequently, parallel conduct allegations must be placed in a context that raises a suggestion of a preceding agreement, not merely parallel conduct that could just as well be independent action.” Mayor & City Council of Balt. v. Citigroup, Inc., 709 F.3d 129, 137 (2d Cir. 2013). “Since mere parallel behavior can be consistent with independent conduct, courts have held that a plaintiff must show the existence of additional circumstances, often referred to as ‘plus’ factors, which, when viewed in conjunction with the parallel acts, can serve to allow a fact-finder to infer a conspiracy.” Apex Oil Co. v. DiMauro, 822 F.2d 246, 253 (2d Cir. 1987). “Such ‘plus factors’ may include, for example: a

common motive to conspire; evidence that the parallel acts were against the apparent economic self-interest of the individual alleged conspirators; or evidence of a high level of interfirm communications.” In re Publ’n Paper Antitrust Litigation, 690 F.3d 51, 62 (2d Cir. 2012) (internal quotation marks and citations omitted). But because “such factors in a particular case could lead to an equally plausible inference of mere interdependent behavior, i.e., actions taken by market actors who are aware of and anticipate similar actions taken by competitors, but which fall short of a tacit agreement,” id., these circumstances “must still lead to an inference of conspiracy.” Citigroup, Inc., 709 F.3d at 137. III. DISCUSSION I will now determine whether a reasonable jury could find that Mylan participated in the

alleged conspiracies related to Bromocriptine, Phenytoin, and Pio Met. I begin, however, with the question whether a reasonable jury could find that Mylan participated in the overarching conspiracy by adhering to the “rules of engagement” that underpin that arrangement, as described in my recent ruling on the overarching conspiracy. Sandoz, Inc., 2025 WL 3470502, at *17 (“[The States] have produced evidence that at least some of the Defendants had at least a tacit agreement that they would conduct business in accordance with the broadly understood ‘rules of engagement,’ which included following any price increase by a rival firm and avoiding ‘poaching’ each other’s customers.”). I address the evidence as to Mylan’s participation in the overarching conspiracy first because it forms the background against which evidence relating to allegations of individual-drug conspiracies can be properly assessed. See Cont'l Ore Co. v. Union Carbide & Carbon Corp., 370 U.S. 690, 699 (1962) (noting that antitrust plaintiffs “should be given the full benefit of their proof without tightly compartmentalizing the various factual components and wiping the slate clean after scrutiny of each” and that “[t]he character and effect of a conspiracy

are not to be judged by dismembering it and viewing its separate parts, but only by looking at it as a whole.”); see also In re Ethylene Propylene Diene Monomer (EPDM) Antitrust Litig., 681 F. Supp. 2d 141, 176 (D. Conn. 2009) (identifying the standard for considering evidence of a conspiracy as “whether, when the evidence [is] considered as a whole, it [is] more likely that the defendants had conspired to fix prices than that they had not conspired to fix prices” (quoting In re High Fructose Corn Syrup Antitrust Litig., 295 F.3d 651, 655 (7th Cir. 2002)). As noted by the States’ expert, Dr. Warren-Boulton, the overarching conspiracy facilitated specific drug conspiracies. ECF No. 962-1 at 478–79 (opining that the overarching agreement was “complementary” with the alleged drug-specific agreements, in that “the drug-specific agreements implement and are facilitated by the overarching agreement.” (emphases omitted)); see also

Sandoz, Inc., 2025 WL 3470502 at *18 (“[T]his commitment to ‘rules of engagement’—this framework for market behavior—created fertile ground for specific agreements on bids, prices, and market shares in individual Drugs at Issue.”). In other words, a reasonable juror could infer that a defendant’s participation in a framework agreement designed to facilitate drug-specific agreements would make the prospect that it colluded as to the pricing and bidding of any particular drug more likely. A. Evidence of Mylan’s Participation in the Overarching Agreement Lubke’s testimony is direct evidence that Mylan adhered to the “rules of engagement” in markets in which it competed with Sandoz. Lubke testified that she and Nesta—who was Mylan’s Vice President of Sales and had authority to approve pricing—arranged to share competitively sensitive information with each other about upcoming price increases, ECF No. 964-1 at 4225, with “an understanding that [she] would use the information not to hurt Mylan,” id. at 4400. The arrangement, Lubke testified, also included an understanding that Mylan and Sandoz would not

“come after” each other’s customers. Id. (“If you come after our customers, we will come after yours.”). Lubke’s description of this arrangement is consistent with the industry “rules of engagement,” which included following competitors’ price increases and avoiding “poaching” each other’s customers. See Sandoz, Inc., 2025 WL 3470502, at *6–7, *17. Lubke also considered Mylan to be a “reasonable” competitor, in the sense that term was often used in the industry to describe a company that “would be following price increases, not punishing a competitor for raising their prices by taking unfair market share, not tanking price just in order to get market share.” ECF No. 964-1 at 4440; see also In re Generic Pharms. Pricing Antitrust Litig., No. 16- CM-27241, 2025 WL 2483981, at *14 (E.D. Pa. Aug. 27, 2025) (discussing July 2013 phone calls between Nesta and Lubke in which Lubke asked Nesta what Mylan's price increases were “so that

[Sandoz] could follow” and Nesta provided that information). Further evidence comes from the adverse inference that may be drawn from Nesta and Sandoz executive Armando Kellum’s invocations of the Fifth Amendment when asked whether Mylan and Sandoz colluded to exchange information, allocate customers and fix prices.8 When

8 The MDL court, before which private plaintiffs are still litigating against the same Defendants as here, declined to draw adverse inferences from Fifth Amendment invocations by Nesta, Aigner, and Kellum because they “have all agreed to withdraw their invocations of the Fifth Amendment and be deposed again.” In re Generic Pharms. Pricing Antitrust Litig., 2025 WL 2483981, at *16. The States point out in their Local Rule 56(a)2 Statement of Facts that Kellum, at least, has not revoked his Fifth Amendment assertion in the District of Connecticut, ECF No. 1019–1 ¶ 29, and Mylan does not suggest that either witness wishes to withdraw. Thus, I assume that Nesta and Kellum continue to invoke the Fifth Amendment in this case. considered alongside other evidence, a witness’s invocation of the Fifth Amendment in a civil case can support an adverse inference that the witness’s testimony would have been unfavorable. Baxter v. Palmigiano, 425 U.S. 308, 318 (1976); LiButti v. United States, 107 F.3d 110, 121–24 (2d Cir. 1997); see also In re Foreign Exch., 2022 WL 294118, *8 (“invoking the Fifth

Amendment privilege against self-incrimination” among “‘plus factors’ in support of finding a conspiracy”). Mylan chiefly argues that a Fifth Amendment invocation is not by itself sufficient evidence to defeat summary judgment. See ECF No. 1001-1 at 9 n.7 (citing In re Mylan N.V. Secs. Litig., 666 F. Supp. 3d 266 (S.D.N.Y. 2023); In re Drummon, No. 24-CV-1668, 2025 WL 1327191, at *8–9 (S.D.N.Y. May 7, 2025) (“an adverse inference drawn from a defendant’s invocation of the Fifth Amendment may not be the sole basis for a finding of liability. Independent, corroborative evidence of wrong-doing must be shown” (internal quotation marks omitted))). But the invocations here do not stand alone; they stand together with Lubke’s testimony and the evidence of contemporaneous communications with competitors, the topics of which may be inferred even without drawing any adverse inferences from Fifth Amendment

invocations. The invocations here are but “one more piece of evidence” in support of Mylan’s participation in the alleged conspiracies. See, e.g., High Fructose Corn Syrup, 295 F.3d at 663. “The way the adverse inference works is that if a witness refuses to answer a question by invoking the Fifth Amendment, the Court can draw an inference that the answer to that question would be adverse to the claimant.”9 In re Drummon, 2025 WL 1327191, at *8. Questions to

9 In its reply brief, Mylan argues that “[t]he States do not identify any particular fact as to which they want the Court to draw an adverse inference, which is necessary before the Court may entertain such a request.” ECF No. 1112 at 8 (citing In re Drummon, 2025 WL 1327191, at *8). In their Joint Opposition to Defendants’ Motions for Summary Judgment, however, the States argue that an adverse inference should be drawn where, “[w]hen asked questions about the alleged pricing and market-allocation agreements, the very individuals who managed pricing and customer allocation decisions (and aligned coconspirators who participated in the same scheme) invoked the which Nesta invoked the Fifth Amendment included, “During those calls, were you sharing competitively sensitive information with Ms. Lubke?”; “And during those calls, were you talking to Ms. Lubke about dividing up customers between Mylan and Sandoz for specific drugs?”; “Did you have a general understanding with Ms. Lubke that Mylan and Sandoz would follow each

other’s price increases?”; and “Did you have an ongoing understanding with Ms. Lubke that Mylan and Sandoz would not compete with each other for customers if they had their fair share of a particular drug market?” ECF No. 964-1 at 5006–07. Kellum invoked the Fifth Amendment in responses to questions about whether he was aware that Lubke “had communications about pricing and customer allocations with competitors,” including Nesta, and that he “followed Mylan’s price increases based on the information that Ms. Lubke provided you. Id. at 3211–12, 14; see also id. at 3213 (“Was there a broad agreement in the generic drug industry to allocate market share consistent with fair share principles?”). When considered alongside Lubke’s testimony and the evidence of contemporaneous communications, a reasonable juror could infer that the answers to these questions would have been unfavorable to Mylan. See Baxter, 425 U.S.

at 318. This evidence is enough to permit a reasonable juror, drawing all inference in favor of the States, to find that Mylan joined the overarching conspiracy by committing to the “rules of engagement.” It is not enough, however, to find that Mylan entered into a conspiracy with respect to any particular drug.10 Adherence to the “rules of engagement,” by itself, is insufficient to find

Fifth Amendment.” ECF No. 1052 at 17. The questions posed to Nesta and Kellum fall into this category. 10 I will consider evidence of a Defendant’s pattern of adhering to the “rules of engagement” even if the Defendant did not sell dermatological drugs. Certainly, evidence that a Defendant implemented the “rules of engagement” in a drug with dermatological applications—like nearly all the Drugs at Issue—is more probative of participation in the overarching conspiracy than evidence that it implemented the rules for non-dermatological drugs, such as those that are the participation in a specific-drug conspiracy because it does not include proof as to any specific product and does not demonstrate injury to the plaintiff. See Publ'n Paper, 690 F.3d at 61 (For a plaintiff “[t]o prevail on a claim of horizontal price fixing . . . the evidence must show that: (1) the defendants conspired to raise prices, and (2) this conspiracy caused injury to the plaintiff in the

form of artificially inflated prices”); see also In re Baby Food Antitrust Litig., 166 F.3d 112, 125 (3d Cir. 1999) (“to survive summary judgment, there must be evidence that the exchanges of information had an impact on pricing decisions”). Such evidence must be coupled with other evidence that Mylan implemented the “rules of engagement” in a particular market. I turn next to the drug-specific evidence involving Mylan. B. Pio Met The States have provided sufficient evidence to raise a genuine dispute about whether Mylan participated in a conspiracy to fix prices and allocate the market for Pio Met. The testimony of Aurobindo’s Thomassey is direct evidence of Mylan’s participation in this alleged conspiracy. Thomassey testified that he colluded with Nesta “[t]o make sure that we split the market share

appropriately and keep the . . . price as high as possible” for Pio Met. SAMF ¶ 12; ECF No. 965- 1 at 459–60. “Because price fixing is a per se violation of the Sherman Act, an admission by the defendants that they agreed to fix their prices is all the proof a plaintiff needs.” High Fructose Corn Syrup, 295 F.3d at 654 (cited by Publ’n Paper, 690 F.3d at 63). Mylan attacks Thomassey’s testimony as inadmissible speculation, citing Delaney v. Bank of Am. Corp., 766 F.3d 163, 169–70 (2d Cir. 2014) (“Materials submitted in support of or in opposition to a motion for summary judgment must be admissible themselves or must contain

subject of this motion. But a pattern of adherence to the “rules of engagement” remains relevant regardless of the pharmacological nature of a given drug. evidence that will be presented in admissible form at trial.”). First, Mylan focuses on Thomassey’s statements that he could not remember the topics of particular calls with Nesta. See, e.g., ECF No. 965-1 at 459 (“Q. What were you talking to Mr. Nesta about on February 8th, 2013? A. I specifically do not remember.” (form objection omitted)). Nor could Thomassey initially recall

whether Mylan conceded Cardinal Health as a customer to Aurobindo or shared any competitive information with Aurobindo. Id. at 461, 609. Mylan contends that Thomassey could recall these facts only when presented with exhibits, such as contemporaneous emails.11 Thomassey was deposed on July 18, 2023. It was not until he was shown call logs reflecting calls between him and Nesta in February 2013 and then asked this question by the States’ counsel, “Given the context that Aurobindo was entering the market for [Pio Met], do you think it’s likely that you were talking to Mr. Nesta about [Pio Met]?”, that he responded, “Yes, definitely it was” and that he was “certain” that his calls with Nesta were “absolutely not social calls.” Id. at 459. It is hardly surprising, however, that Thomassey needed his memory refreshed with respect to phone calls that had taken place over a decade earlier.12

Mylan further objects that the States’ questions were leading. While I agree that the States’ “[g]iven the context” question was leading, leading questions may be used to refresh recollection and, in any event, on summary judgment evidence need not be presented in admissible form as

11 As I explained in the overarching conspiracy ruling, the deposition transcripts of Thomassey in the summary judgment record raise no concerns about his memory or his competence as a witness. See Sandoz, Inc., 2025 WL 3470502, at *6 n.6. The States’ renewed Daubert motion to exclude the opinions of Dr. Levin for the purposes of Group 3 summary judgment motions, ECF No. 911, is denied without prejudice for the same reasons I explained previously. See id. 12 Thomassey testified that he had a “strong core memory of the conduct that [he] engaged in with specific competitors.” ECF No. 965-1 at 612. I disagree with Mylan that this statement related only to Thomassey’s prior conduct while he worked for Fougera, and not his conduct at Aurobindo. See ECF No. 1112 at 14. In any event, a reasonable juror could infer that Thomassey communicated with competitors for similar purposes at both companies. long as the evidence can be presented in admissible form. Figueroa v. Mazza, 825 F.3d 89, 98 n. 8 (2d Cir. 2016) (“any evidence considered on summary judgment must be reducible to admissible form”); Bryant v. Farms Ins. Exchange, 432 F.3d 1114, 1122 (10th Cir. 2005) (“At the summary judgment stage, the parties need not submit evidence in a form admissible at trial; however, the

content or the substance of the evidence must be admissible.”). The substance of his response to the leading question is plainly admissible. Further, after being shown more documents, Thomassey confirmed that he had reached an agreement with Nesta in response to non-leading questions as well. See ECF No. 965-1 at 461–62 (“Q: Did Mylan concede Cardinal to Aurobindo as it was entering the market for [Pio Met]? A: Yes. Q: And did that have anything to do with the conversations you were having with Jim Nesta at Mylan? A: Yes. Q: Was there an agreement between Aurobindo and Mylan regarding [Pio Met]? A: Yes. Q: And what was that agreement? A: That they would walk and we would . . . get the award at Cardinal.” (objections omitted)). To the extent this testimony contradicted other answers he gave about his inability to recall particular calls specifically, it is for a jury to resolve any such contradictions. And I disagree

with Mylan’s claims that this and similar testimony by Thomassey was not based on his personal knowledge. Thomassey may not have remembered the contents of specific calls, but the States’ call logs showed that he participated in the calls and Thomassey had personal knowledge of his own business practices, e.g., after calls with a competitor, he would reach out to his boss to discuss bids and pricing, after which his employer would submit a bid that was more or less in line with prevailing prices in the market. The pattern he testified to concerning his communications with Nesta closely tracks the one he described repeatedly when dealing with other competitors. See Connecticut v. Sandoz, 2025 WL 3470502, at *8–9. In short, the substance of his testimony is admissible, and it will be for a jury to decide how much weight, if any, to accord to it. Publ’n Paper, 690 F.3d at 64 n.8 (“Questions about the weight to be given to testimony based on the witness’s “credibility, perception, or memory are, of course, for the jury.”). In any event, Thomassey’s testimony does not stand alone. The volume and timing of frequent interfirm communications also contribute to a finding that it is more likely than not that

Mylan participated in a Pio Met conspiracy. Specifically, as Mylan’s period of exclusivity neared its end, Mylan’s Nesta communicated with potential new entrants to the Pio Met market in January 2013, and he called Thomassey at Aurobindo immediately after writing an internal email about the importance of obtaining competitors’ information in efforts to maintain pricing. SAMF ¶ 2 (noting twenty-six calls with Aurobindo, Teva, and Ranbaxy); ECF No. 959-1 at 188. Nesta’s calls with competitors, including Thomassey, continued as other companies entered the market in February. SAMF ¶¶ 5–6. Although the phone records alone do not show the topics of discussion, a reasonable juror could infer from Thomassey’s sending of emails about Aurobindo’s pursuit of Cardinal’s Pio Met business soon after two calls with Nesta on February 8 and 11, 2013, that the calls with Nesta

pertained to whether Mylan or Aurobindo would serve as Cardinal’s supplier. Id. ¶¶ 6, 8. Further, during a single afternoon on February 8, 2013, Thomassey exchanged seven calls with Nesta and nine calls with his boss, Grauso, including a final, 24-minute call during which he also emailed Aurobindo’s Pio Met offer to Cardinal. Id. ¶ 6; ECF No. 951-1 at 65. The inference that these communications concerned Pio Met is reinforced by Nesta’s email two days later, in which he opined that “[w]e should be good to go on [] Pio Met” because he had “locked all this up last week.” SAMF ¶ 7. There is also evidence that Thomassey and Grauso were speaking to Teva and Sandoz about allocating the Pio Met market at around the same time Thomassey was speaking to Nesta. Thomassey and Sandoz’s Bihari spoke three times on February 19, 2013, and Bihari's contemporaneous notes show that he and Thomassey exchanged pricing information, which Bihari testified “allowed Sandoz to appropriately price [Pio Met] in the market upon launch and to target these customers.” Id. ¶ 15; ECF No. 964-1 at 920. This exchange was pursuant to an agreement

to “ensure Sandoz keeps their pricing high in the market upon entry and to target certain customers.” ECF No. 964-1 at 920. Thomassey’s boss, Grauso, exchanged frequent calls between January 28, 2013, and February 13, 2013 with Teva’s Green and another Teva employee while pursuing another customer, Harvard. SAMF ¶ 14. Although these communications do not directly implicate Mylan, they make a market-wide agreement for Pio Met more likely. It is reasonable to infer that if Aurobindo had an agreement with Sandoz or Teva (or both) about Pio Met, it was more likely to have a similar agreement with Mylan, with whom Thomassey also frequently communicated during this time, to ensure that as many Pio Met suppliers as possible were on the same page. Also supporting this inference is the fact that Thomassey and Grauso expressed frustration at Torrent for failing to maintain high prices, which stands in contrast to their

communications about other competitors. Nesta also demonstrated familiarity with the “rules of engagement” that form the basis of the overarching conspiracy, including that all competitors in a given market are entitled to their “fair share.” Nesta wrote on May 20, 2013, that Mylan must “let go” of Publix’s Pio Met business because Sandoz was a “[n]ew entrant. We have 40 share.” Id. ¶ 18. Although he did not use the term “fair share,” a reasonable jury could infer that Nesta was describing the concept—when a new entrant entered the market, Mylan would have to give up share. Mylan is correct that this statement, standing alone, is equally as probative of independent conduct as it is of collusion. But it does not stand alone. As I found in the overarching conspiracy ruling, an understanding of the rules of engagement, including the allocation of fair share, “facilitated the consummation of specific agreements on price and customer allocation for individual Drugs at Issue . . . [and] made it easier to reach those agreements as new bidding opportunities arose.” Sandoz, Inc., 2025 WL 3470502, at *17. In this instance, Nesta called Lubke at Sandoz later on May 20, 2013. SAMF ¶

18. A reasonable juror could infer that this call, and others between Nesta and Lubke in this period, were part of discussions between Mylan and Sandoz for Mylan to cede Publix to Sandoz to maintain high prices in the market. Moreover, Teva’s Green testified that he and Nesta frequently exchanged information about whether Mylan would be “holding on to a lot of their share” when Teva entered a drug market.13 Id. ¶ 21. As this is the same type of exchange that characterizes the overarching conspiracy, it also supports the inference that when Nesta and Green spoke in February 2013 they were implementing the rules of engagement during their discussions of Pio Met to allocate the market and maintain prices. As with the other drugs, Nesta and Kellum’s Fifth Amendment invocations also support

the inference that Mylan participated in a Pio Met conspiracy. Id. ¶¶ 22–23. Nesta invoked the

13 In its reply brief, Mylan argues that Green’s testimony is inadmissible because it was completed after the close of discovery in this action. ECF No. 1112 at 13 n.9. Soon after this case and the other two State AG actions against generic drugmakers were assigned to me, the parties filed a joint status report that indicated that, while fact discovery had closed in this action, fact depositions were continuing for Defendants who were only named in the other two actions, including Teva. ECF No. 156 at 7–8. The parties also noted that, “Discovery in the MDL was consolidated, such that any discovery was deemed to be taken in all of the cases. Thus, all of the fact discovery described above is deemed to have been taken not only in the Dermatology Action, but also in the Teva and Heritage Actions. Certain written discovery requests and certain depositions remain outstanding.” Id. at 12. Given these unique circumstances, and because Green could testify to these same facts at trial, I will consider Green’s testimony at summary judgment even though it was given after the close of fact discovery in the Dermatology action. See Delaney, 766 F.3d at 169–70 (“Materials submitted in support of or in opposition to a motion for summary judgment must be admissible themselves or must contain evidence that will be presented in admissible form at trial.” (emphasis added)). Fifth Amendment when asked whether Mylan and Teva had an agreement to allocate the Pio Met market during the 180-day exclusivity period, and whether he agreed with Teva’s Green to divide the market between their companies. ECF No. 964-1 at 5146. Nesta also invoked the Fifth Amendment in response to questions about pricing, market share goals, and dividing the Pio Met

market in calls with Aurobindo’s Thomassey in January and February 2013, and with Sandoz’s Lubke in April 2013. Id. at 5148–49, 5151. Nesta again invoked the Fifth Amendment when asked, “Did you reach an agreement with Anthony Thomassey and Kevin Green about how much market share each company would get for pio met?,” in response to, “Did you agree that Mylan would concede other customers to Aurobindo for pio met?”, and in response to “Was there an agreement between Mylan, Teva and Aurobindo to minimize competition and keep prices as high as possible for pio met?” Id. at 5150–51. Kellum invoked the Fifth Amendment when asked if Sandoz had an understanding with its Pio Met competitors “that Sandoz would win its market share from Teva and Mylan and not Aurobindo or Torrent?” and when asked whether each competitor, including Mylan, agreed to fix prices and allocate customers. Id. at 3385–86. When

considered alongside the testimony of Thomassey and the other cooperating witness, as well as the evidence of contemporaneous communications, a reasonable juror could infer from these invocations that responses to these questions would have been unfavorable to Mylan, including about whether Mylan had agreements with Teva and Sandoz to fix prices and allocate the market. See Baxter, 425 U.S. at 318. With all of this evidence pointing toward a conspiracy, the fact that Pio Met prices fell during the alleged period of anticompetitive conduct is not enough to make it more likely than not that changes in the market were a result of independent actions, or that no Pio Met conspiracy existed. As described above, the evidence suggests that price erosion in the Pio Met market followed the entrance of Torrent, which did not play by the other competitors’ “rules of engagement” or communicate sensitive information, much to the others’ chagrin. See ECF No. 965-1 at 575 (Thomassey referring to Torrent as “morons” because they were “so bad at eroding prices”). As the States’ expert, Dr. Warren-Boulton, opined, in a market share agreement

(“MSA”), “any one firm can move the entire set of firms to any higher price, as long as all firms comply with the terms of the MSA.” ECF No. 952-1 at 128. In other words, if one firm does not comply, its competitors cannot move to a higher price because the new price will always be undercut by the noncompliant firm, at least to the extent that the noncompliant firm has the capacity to supply additional customers. Thus, the price of a drug alone is not evidence against a conspiratorial inference where the market included any firm that did not participate in the conspiracy. Nor is price erosion evidence that the other competitors did not form agreements among themselves to allocate the market and keep prices high. A conspiracy to restrain trade does not need to be successful to violate the Sherman Act. See High Fructose Corn Syrup, 295 F.3d at 656 (courts must “distinguish between the existence of a conspiracy and its efficacy”); United

States v. Foley, 598 F.2d 1323, 1333 (4th Cir. 1979) (“Since the agreement itself, not its performance, is the crime of conspiracy, the partial non-performance of [a defendant] does not preclude a finding that it joined the conspiracy.”). Nor, contrary to Mylan’s assertions, do Aurobindo’s initial movements in the Pio Met market suggest the absence of collusion. As Dr. Warren-Boulton wrote, when first entering market, an “entrant manufacturer can be expected to set a price that is initially slightly lower than the incumbent suppliers’ prices so as to acquire its fair share with minimal price erosion.” ECF No. 952-1 at 134. If a new entrant did not offer a lower price, the customer would have no incentive to switch away from their incumbent supplier. Id.; see, e.g., ECF No. 958-1 (Thomassey writing in 2012 that a new offer needed to be twenty percent less to “incentivi[z]e a switch”). Aurobindo’s bid suggests that the January 2013 calls between Mylan and Aurobindo included discussion of the price point that would incentivize Cardinal to switch suppliers while still minimizing price erosion. See ECF No. 952-1 at 134–35 (Warren-Boulton: “This initial price maneuvering or customer

allocation requires communication between participants so as to minimize disruption. . . . Information on which customers the incumbent is willing to relinquish and the prices currently being charged to those customers allows the entrant to submit a ‘competitive’ offer to the designated customers that is just low enough to trigger a right of first refusal by the incumbent, secure in the knowledge that the incumbent will not respond to a bid request from those customers.”) Thus, it is likely that by communicating with Mylan, Aurobindo was able to offer a higher entry price than if it had acted independently. Finally, Mylan argues that there is no evidence that the alleged conduct caused any injury, because even if Mylan and Aurobindo conspired in setting Aurobindo’s offer price to Cardinal, Torrent drove prices lower before the collusion-related pricing went into effect. See ECF No.

1019-1 ¶¶ 96–98. To avoid summary judgment, the States “must prove the existence of antitrust injury, which is to say injury of the type the antitrust laws were intended to prevent and that flows from that which makes [Mylan’s] acts unlawful.” Warner-Lambert Co. v. Schick U.S.A., Inc., 935 F. Supp. 130, 144 (D. Conn. 1996) (quoting Atlantic Richfield Co. v. USA Petroleum Co., 495 U.S. 328, 334 (1990)). If the price Cardinal ultimately paid Aurobindo was unrelated to prior collusion, the States would not be able to show antitrust injury. The evidence shows that Aurobindo initially contracted with Cardinal to supply Pio Met at a contract price of $143.95, and after Torrent entered the market at $90, Aurobindo offered a reduced price of $115.30, which Cardinal accepted. ECF No. 1019-1 ¶¶ 90–99. This raises the question of why Cardinal would remain with Aurobindo at a price point that was twenty-two percent higher than might have been available from Torrent. Considering the evidence in the aggregate, a reasonable juror could infer that this difference was not enough to incentivize a switch to Torrent from Aurobindo, with which Cardinal was already contracted. In

other words, when drawing all reasonable inferences in the States’ favor, a reasonable juror could find that Aurobindo’s contract with Cardinal, which began at a conspiracy-related price, still impacted the price Cardinal paid. Although Aurobindo discounted the ultimate price following Torrent’s entrance, a reasonable juror could find that it still stemmed from and was affected by Aurobindo’s initial price. In a “but-for” world in which Aurobindo had not colluded, its initial price might have been lower than the price Cardinal ultimately paid. See ECF No. 952-1 at 206- 07 & fig.32 (Dr. Warren-Boulton calculating the “but-for” price for Pio Met 850 mg below the defendant net price throughout the conduct period). Therefore, I cannot find on the summary judgment record before me that Cardinal suffered no antitrust injury. When considering the testimony of cooperating witnesses, the evidence of

contemporaneous phone calls and notes, analysis of the States’ expert, and adverse inferences from Nesta and Kellum’s Fifth Amendment invocations, a reasonable jury could find that Mylan participated in a conspiracy with at least one competitor to set prices and allocate the Pio Met market. Thus, the motion for summary judgment on the States’ Pio Met claims is DENIED. C. Bromocriptine The States have provided sufficient evidence to raise a genuine dispute of fact that Mylan participated in a conspiracy to fix prices for Bromocriptine. This includes direct evidence from Lubke’s testimony that she discussed “competitive intelligence” with Mylan’s Nesta in March 2013, around the same time that Mylan exited the market due to supply issues; that it was “probable” that they discussed Bromocriptine supply issues on those calls; and that these discussions were “consistent” with their information-sharing arrangement. ECF No. 964-1 at 4325.14 But the mere exchange of competitively sensitive information does not prove a per se violation of the antitrust laws. See United States v. Citizens & Southern National Bank, 422 U.S.

86, 113 (1975) (holding, in a case where the trial court found no collusive price fixing, that “the dissemination of price information is not itself a per se violation of the Sherman Act”). Thus, the States must also point to circumstantial evidence that these exchanges of information led to anticompetitive collusion. See In re Generic Pharms. Pricing Antitrust Litig., 2025 WL 2483981, at *14 (“The question is whether the phone logs, notes, and deposition testimony together create a basis by which a reasonable factfinder could conclude that Mylan, Sandoz, and Taro developed a common plan and made pricing decisions based on those conversations.”) Foremost among the States’ circumstantial evidence is Mylan’s exchange of phone calls with competitors that were contemporaneous with price increases, especially Nesta’s ten calls with Lubke over a 24-day period leading up to Mylan’s reentering the Bromocriptine market at the

Sandoz price, which had sustained an increase of over 200 percent only a month earlier. When considered with evidence of parallel conduct, a high level of interfirm communications within an

14 Mylan’s attempts to minimize Lubke’s testimony are unconvincing. While Lubke later testified in response to defense counsel’s question that she could not specifically recall speaking to Nesta about Bromocriptine, ECF No. 964-1 at 4401, her statement that Bromocriptine was a probable topic of discussion is enough to raise a dispute of fact. Moreover, in this line of questioning, Lubke was asked about only two calls with Nesta in March 2013, not about the ten phone calls she exchanged with Nesta in May 2013. See id.; SAMF ¶¶ 16, 25. So Mylan’s assertion that Lubke “had no recollection of ever talking to Nesta about Bromocriptine,” ECF No. 1001-1 at 16 (emphasis added), is not supported by the deposition transcript. The fact that a Bromocriptine supply issue was already “old news” to Mylan is irrelevant to determining whether Lubke and Mylan discussed fixing the price of Bromocriptine on any of their calls. Finally, although Lubke did not have pricing authority herself and did not make commitments on behalf of Sandoz, she testified that if she received competitive intelligence from Nesta about Bromocriptine, she would have reported it to her superiors. ECF No. 964-1 at 4325. oligopoly can be a critical plus factor supporting the inference of an agreement between competitors. See Valspar Corp. v. E.I. Du Pont De Nemours & Co., 873 F.3d 185, 193 (3d Cir. 2017) (focusing on “evidence implying a traditional conspiracy,” which requires “proof that the defendants got together and exchanged assurances of common action or otherwise adopted a

common plan even though no meetings, conversations, or exchanged documents are shown”). Naturally, the more frequent and the closer the calls are to a particular movement in the market, the more probative of a conspiracy this evidence becomes. See In re Flat Glass Antitrust Litigation, 385 F.3d 350 (3d Cir. 2004) (“So, for example, there is evidence that AFG faxed to PPG a copy of a planned future increase that it had not announced publicly, PPG announced an identical increase before AFG, and the rest of the flat glass producers followed with identical price increases. . . . In sum, here the exchanges of information are more tightly linked with concerted behavior and therefore they appear more purposive. . . . The inference of concerted rather than interdependent action is therefore stronger.”); EPDM, 681 F. Supp. 2d at 173 (denying summary judgment in Section 1 case where plaintiffs submitted documents showing “that the defendants

were extremely chummy and treated each other more like colleagues than traditional competitors and that the defendants’ executives often communicated immediately before or after [a] price increase.”); In re Urethane Antitrust Litig., 913 F. Supp. 2d 1145, 1155 (D. Kan. 2012) (denying summary judgment in Section 1 case where circumstantial evidence included “evidence of a great number of communications and meetings and even vacations involving executives for these competitors, including communications involving pricing, and including communications at or near the time of price increase announcements by the companies”). Here, Nesta spoke with Sandoz’s Lubke twice on May 23, 2013—the same day on which Mylan’s internal pricing team recommended following Sandoz’s Bromocriptine price increase, which Nesta approved. SAMF ¶¶ 25, 30. Nesta also spoke with Lubke four times on May 31, 2013, the day that Mylan’s price increase went into effect. Id. ¶ 25. Although the record does not include the specific content of these calls, the proximity to Mylan’s price increases, together with Lubke’s testimony that she had “probabl[y]” discussed Bromocriptine with Nesta back in March

and that every call she had with him was to discuss competitive intelligence (as well as Nesta’s subsequent approval of the price increase) is sufficient to permit a reasonable juror to infer that the calls included discussion of whether and how Mylan would follow Sandoz’s Bromocriptine price increase.15 See In re Generic Pharms. Pricing Antitrust Litig., 2025 WL 2483981, at *14 (finding sufficient evidence to establish the interfirm communications plus factor where “calls occurred immediately before or after pricing decisions, and that those calls occurred in tight temporal proximity with calls to and between other representatives”). And because Mylan was able to “recapture[]” many of its previous Bromocriptine customers at the higher price once it reentered the market, ECF No. 1019-1 ¶ 23, a reasonable juror could infer that the May 31 calls included discussion about how to allocate the market.16

I agree with Mylan that, when viewed in isolation, the mere act of following Sandoz’s price

15 Even if some of Lubke and Nesta’s ten calls in May 2013 were about Pio Met, and not Bromocriptine, see SAMF ¶ 18, a reasonable juror could find that it is unlikely that all of the calls that coincided with movement in the Bromocriptine market pertained to a different drug. 16 I reject Mylan’s argument that its price increase was not “parallel” with Sandoz’s because it came two months later. See ECF No. 1001-1 at 13–14 (citing In re Concrete & Cement Additives Antitrust Litig., No. 24-md-3097 (LJL), 2025 WL 1755193, at *12–13 (S.D.N.Y. June 25, 2025) (“The relative timing of supposedly parallel conduct is particularly critical . . . even highly similar actions separated by a lengthy delay are not truly parallel.”). That same ruling said that “there is no categorical rule governing how temporally proximate acts must be to give rise to an inference of collusion,” and that pricing changes must be “viewed in the context of the case” and “must account for any restrictions on market participants’ ability to enact price changes sooner.” Id. at *13 n.11. Here, when Sandoz increased it price, Mylan was out of the market because of supply issues. Mylan then followed Sandoz’s increase as soon as it reentered. Viewed in this context, a reasonable juror could find that the price increases demonstrated parallel conduct. increase was a logical step for Mylan that was at least equally suggestive of independent conduct as collusion. Mylan was losing money on Bromocriptine before the conduct period began, and likely could not seek to expand its market share because of supply problems. But just as I found in the overarching conspiracy ruling, “[t]he problem is that the States have submitted evidence that

the Defendants communicated with each other . . . to provide assurance that each would follow” price increases to avoid competing with each other. Sandoz, Inc., 2025 WL 3470502, at *20. Communications with competitors are not, without more, illegal. But because these communications correspond with changes in the competitors’ positions in the market, they support an inference that Mylan colluded with Sandoz to fix prices and allocate the market. See In re Generic Pharms. Pricing Antitrust Litig., 2025 WL 2483981, at *14. Adding to this evidence is the testimony from other Sandoz employees, Bihari and Vezza, that Bihari was colluding with his counterpart at Perrigo to fix the price of Bromocriptine and other Drugs at Issue during this time. Bihari testified that it was unusual for Sandoz to lead a price increase, and that to do so the company must have had a “high level of confidence that Perrigo

would follow and not take their customers.” ECF No. 964-1 at 922. Bihari also testified that Perrigo, while slow to follow a price increase, also provided assurances to Sandoz that it would not poach Sandoz’s customers while it was offering a lower price. SAMF ¶ 33. Although there is no evidence that Bihari spoke with anyone at Mylan, the testimony that Sandoz sought assurances from and had already conspired with its top competitor on Bromocriptine suggests at least that Sandoz had a compelling motive to conspire with Mylan in the same market to avoid a situation in which Mylan might disrupt the plans of Sandoz and Perrigo. And the calls between Bihari’s colleague, Lubke, and Nesta in March 2013—shortly after Mylan customer Walgreens asked if Sandoz could temporarily supply the drug and shortly before Sandoz’s price increase—and again in May 2013, would allow a reasonable juror, drawing inferences against Mylan, to find that Sandoz, just as it did with Perrigo, sought assurances that Mylan would follow a price increase when the latter reentered the Bromocriptine market and that Sandoz and Mylan would not poach each other’s customers.17

The circumstantial evidence also includes Nesta’s and Kellum’s invocations of the Fifth Amendment in response to questions about Bromocriptine during their deposition testimony. In addition to the general questions about sharing competitively sensitive information, Nesta also invoked the Fifth Amendment in response to questions about Bromocriptine, including, “Did you tell Ms. Lubke that Mylan would follow Sandoz’s price increase on bromocriptine mesylate?”; “Did you reach an agreement with Ms. Lubke about which customers Sandoz would concede to Mylan as it re-entered the market for bromocriptine mesylate tablets?”; “Did Mylan have an agreement with Perrigo and Sandoz to raise the prices of bromocriptine mesylate tablets?”; and “Did Mylan have an agreement with Perrigo and Sandoz to divide the market for bromocriptine mesylate tablets?” ECF No. 964-1 at 5015–21. Kellum also invoked the Fifth Amendment in

response to questions about Bromocriptine, including, “Would you have implemented a price increase on that product if there wasn't an agreement from Perrigo and Mylan to follow that price increase?”; “Did you have an understanding at this point in time that Mylan would follow this price increase when its supply issue resolved?”; and “Was it your expectation that Ms. Lubke would also be taking steps to confirm Mylan’s commitment to the agreement on bromocriptine?” Id. at 3322–25. When considered alongside Lubke’s testimony and the evidence of

17 Mylan also points out that Bihari and Vezza denied that they had or knew of any agreement with Mylan. ECF No. 1019-1 ¶¶ 31, 38. These denials, however, do not refute the evidence that Lubke had an agreement with Mylan, or that she reported competitive intelligence to Kellum, who was senior to her, Bihari, and Vezza. Thus, denials from Bihari and Vezza do not rule out any agreement between Sandoz and Mylan. contemporaneous communications, a reasonable juror could infer from these invocations that responses to these questions would have been unfavorable to Mylan, including about whether Mylan had agreements with Perrigo and Sandoz to fix prices and allocate the market for Bromocriptine.18 See Baxter, 425 U.S. at 318.

When considered together, the evidence presented—Lubke’s testimony regarding Mylan’s understanding of the “rules of engagement” and general conversations with Nesta, together with her testimony that she “probabl[y]” spoke with him about Bromocriptine; the phone records of communications between competitors contemporaneous with price changes; and adverse inferences from Nesta’s and Kellum’s Fifth Amendment invocations—“tends to exclude the possibility” that Mylan “acted independently.” Matsushita Elec. Indus. Co., 475 U.S. at 588. Because the States have raised a genuine dispute of fact as to whether Mylan joined the alleged Bromocriptine conspiracy, Mylan’s motion for summary judgment on the States’ Bromocriptine claims is DENIED. D. Phenytoin

The States have also provided sufficient evidence to raise a genuine dispute of fact that Mylan participated in a conspiracy to fix prices and allocate the market for Phenytoin. With Phenytoin, the States do not present any direct evidence from cooperating witnesses but rely entirely on circumstantial evidence of parallel conduct, paired with numerous interfirm communications. The frequency and timing of these communications, however, would permit a reasonable juror to infer that Mylan agreed with at least one competitor to follow price increases

18 I agree with Mylan that the States have not produced evidence—other than the participants’ invocations of the Fifth Amendment—that the single September 2014 call between Nesta and Polman related to Bromocriptine or was contemporaneous with any changes in the Bromocriptine market. I therefore do not rely on that call as evidence of Mylan’s participation in a Bromocriptine conspiracy. and avoid poaching each other’s customers. To start, a reasonable juror could find that Taro’s decision to double its price in a five-firm market was “not economically justified or supportable, but required competitors to hold the line.” Flat Glass, 385 F.3d at 369. As discussed in the overarching conspiracy ruling, “if a supplier

declines to follow an initial price increase in an attempt to increase its market share, it risks triggering a ‘race to the bottom’ in which prices erode as suppliers compete for customers.” Sandoz, Inc., 2025 WL 3470502, at *4. For Taro’s large price increase to “stick,” it needed each of its competitors to follow—rather than undercut—its new price. The volume of interfirm communications during this period supports the inference that the competitors were seeking assurances that each competitor would be amenable to following a price increase. On April 7, 2014, Taro listed Phenytoin among a number of drugs for which it wanted to increase prices. ECF No. 952-1 at 54–58. Three days later, phone records show, Taro’s Aprahamian contacted a Phenytoin competitor, exchanging two phone calls with Mylan’s Aigner. SAMF ¶ 39. The records also show that Mylan’s Nesta and Amneal’s Rivero exchanged eight

phone calls and twenty-six text messages between April 22 and April 30. That same month, Aprahamian and Mylan’s Nesta, as well as Amneal’s Rivero, all attended an industry event, during which Aprahamian requested Phenytoin pricing information from a colleague, and after which an internal Mylan document indicated a “Potential Amneal price increase” for Phenytoin. Id. SAMF ¶¶ 38, 40; ECF No. 952-1 at 60. No further communication occurred between Mylan and Taro before Taro increased its Phenytoin price on June 2, 2014. ECF No. 1019-1 ¶ 64. Although the States do not provide evidence of the topic of the April calls or of any discussion at the trade meeting, when viewed in the context of Taro’s contemplating a major price increase on Phenytoin, and when drawing all inferences in favor of the States, one reasonable inference is that Aprahamian sought information from Taro’s competitors about how they would react to a price increase, and that Taro’s competitors subsequently discussed the matter among themselves. By itself, however, this evidence is insufficient to find that it was more likely than not that the competitors conspired to fix prices or allocate the market for Phenytoin. But the communications did not stop there.

After Taro doubled its price on June 2, 2014, the immediate concern voiced by Amneal’s Rivero was that she hoped that “Mylan sees this,” after which her colleague Lowther opined that Mylan would follow. SAMF ¶ 45. From this evidence, a reasonable juror could infer that Amneal was concerned about whether Mylan would follow Taro’s increase. On the same day and in the days that followed, both Rivero and Lowther exchanged calls with Mylan employees. Id. ¶ 44. Two days after Taro’s increase, Lowther emailed Rivero a document titled “Phenytoin Price Increase” that “[m]atched the new Taro” prices. ECF No. 968-2 at 722. On June 18, Nesta internally approved Mylan’s price increase. One week later, Lowther wrote in an email, “We are matching Taro’s new [price] and expect Mylan to do the same.” ECF No. 964-1 at 6269. From this evidence, a reasonable juror could infer that Amneal sought and received assurances from

Mylan that it would follow Taro’s increase. The communications continued as Mylan and Amneal implemented their price increases. Nesta spoke nineteen times with Amneal contacts in June 2014, both before and after he approved Mylan's price increase. Meanwhile, Aigner was also speaking to Taro’s Aprahamian in June and July, just before Mylan announced it was following Taro’s increase on July 16, 2014. Amneal and Sun also followed the price increase in July, meaning that all Phenytoin competitors increased their prices with six weeks of each other. Although the Phenytoin competitors did not increase their prices in unison, their price increases were close enough to demonstrate parallel conduct. See Mosaic Health, Inc. v. Sanofi-Aventis U.S., LLC, 156 F.4th 68, 82 (2d Cir. 2025) (timing was “similar” for purposes of finding parallel conduct where three of four Defendants announced changes within one month of each other). Thus, even if Taro’s initial increase, in the absence of more substantial prior communications with Phenytoin competitors, could be seen as independent conduct, the extensive

communications evidence from June and July 2014 would permit a reasonable juror to find that Mylan’s decision to follow Taro’s increase was more likely than not the product of concerted action. See Flat Glass, 385 F.3d at 369 (“[T]he exchanges of information are more tightly linked with concerted behavior and therefore they appear more purposive. Several of the key documents . . . suggest not just foreknowledge of a single competitor’s pricing plans, but of the plans of multiple competitors. Predictions of price behavior were followed by actual price changes. The inference of concerted rather than interdependent action is therefore stronger.”). The June and July 2014 communications, which occurred while Mylan’s pricing team was analyzing whether following Taro would be profitable and, later, while preparing to announce price increases to customers, go well beyond “sporadic exchanges of shop talk among field sales representatives who

lack pricing authority.” Baby Food, 166 F.3d at 125, and permit an inference that Mylan and Amneal reached an agreement that they would both follow Taro’s increase rather than compete for each other’s customers at a lower price.19 The calls between Lowther and Mylan employees also tend to exclude the possibility that Lowther was merely “making an assumption” about Mylan’s plans to follow Taro’s increase, as Rivero testified, and that, instead, Mylan had told him it would follow Taro. ECF No. 964-1 at 6270.

19 Even if Mylan conducted its profitability analysis independently, a reasonable juror could find that Mylan did not use this analysis exclusively in deciding whether to follow Taro’s price increase. To survive summary judgment, the States need not rule out every alternative explanation for why Mylan followed Taro’s increase; they only need to show that a conspiratorial explanation is more likely than not. Anderson News, 899 F.3d at 99. When considering the evidence in total, including the evidence that led me to find that Taro stood among “a core group of leaders most responsible for perpetuating the Defendants’ collective adherence to the rules of engagement,” Sandoz, Inc., 2025 WL 3470502, at *21, a reasonable juror could infer that Mylan’s calls during this period were made to determine whether and how Mylan

would follow Taro’s increase. In sum, the volume of interfirm communications and their proximity to major changes in prices tend to exclude the possibility that Mylan acted independently in choosing to follow Taro’s price increase. Mylan contends that no inference of parallel conduct may be drawn from sequential price increases where each competitor learned about increases through public sources. ECF No. 1001- 1 at 19 (citing In re Chocolate Confectionary Antitrust Litig., 801 F.3d 383, 408–09 (3d Cir. 2015) (“gathering the price information of competitors can be just as consistent with lawful interdependence as with a price-fixing conspiracy”)). In Chocolate Confectionary, however, the Third Circuit noted that the information that the competitors shared did “not reveal pricing plans dependent on others following,” and that competitors’ “pricing actions were intended to, and in

some cases did, catch their rivals by surprise.” 801 F.3d at 408. That scenario bears no resemblance to the facts here, which include evidence that competitors communicated with each other frequently immediately before deciding to follow a price increase, together with evidence that competitors were hesitant to lead price increases without confidence that competitors would follow. See, e.g., ECF No. 965-1 at 964 (Vezza testifying that, if Sandoz believed in advance that a competitor would not follow its price increase, “we would avoid. We would not take that price increase.”). In addition to the communications evidence, I also consider that Mylan’s Aigner and Nesta, as well as Taro’s Aprahamian, all invoked the Fifth Amendment when asked about Phenytoin- related discussions with competitors. Aigner invoked the Fifth Amendment in response to questions about whether he discussed Mylan’s Phenytoin bid to Amneal’s customer Walgreens during communications with Amneal’s Rivero in May 2014, including seven phone calls and one text message on May 29—one day before Mylan submitted its bid. ECF No. 964-1 at 43–44. He

also invoked the Fifth Amendment in response to questions about whether he reached any agreements with Amneal’s Rivero or Taro’s Aprahamian in June 2014 regarding pricing or market share for Phenytoin, as Mylan was deciding whether to follow Taro’s price increase. Id. at 45. And he invoked the Fifth Amendment in response to a question about whether he informed Rivero in July 2014 that Mylan “was no longer going to go after Walgreens’ phenytoin sodium ER business that was then with Amneal.” Id. at 46. He again invoked the Fifth Amendment in response to questions about whether Mylan agreed with either Taro or Amneal to fix or raise Phenytoin prices or allocate Phenytoin customers. Id. at 47. Nesta and Aprahamian, as they did in response to questions about Bromocriptine, also invoked the Fifth Amendment in response to questions about Phenytoin. Nesta, for instance,

invoked the Fifth Amendment when asked if he had “reach[ed] an agreement with Shannon Rivero that if . . . Amneal raised its pricing on Phenytoin sodium ER capsules that Mylan would follow that price increase[.]” Id. at 5141. Nesta also invoked the Fifth Amendment in response to questions about whether there was an agreement between Mylan, Amneal, and Taro to raise prices or allocate the market for Phenytoin. Id. at 5145. Similarly, Aprahamian invoked the Fifth Amendment in response to questions about whether Taro had agreements with either Amneal or Mylan that resulted in higher Phenytoin prices. Id. at 242. These invocations, when considered alongside the evidence of contemporaneous interfirm communications, would permit a reasonable juror to infer that responses to these questions would have been unfavorable to Mylan, including about whether Mylan had agreements with either Taro or Amneal to fix prices and allocate the Phenytoin market. See Baxter, 425 U.S. at 318. The expert reports provide further support for denying summary judgment here. The States’ expert, Dr. Warren-Boulton, opined that “the primary indicator of collusion is a large

increase in price that is not explained by a change in market or structural conditions,” ECF No. 952-1 at 189, and that relatively stable market shares following an increase are indicative of collusive behavior as opposed to the “cautious process of step by step signal and response” typically observed in oligopolies, ECF No. 960-1 at 37. As noted above, the Phenytoin market saw a large price increase followed by stable market shares, aside from Amneal’s initial gain of some share at Taro’s expense. While Dr. Warren-Boulton’s opinions do not by themselves tend to exclude the possibility of independent conduct, when considered with the other evidence above, they also support an inference that the competitors colluded by agreeing to follow Taro’s increase. Meanwhile, the opinion of Mylan’s expert, Dr. Stiroh, that the Phenytoin market share is inconsistent with an agreement to allocate the market is at odds with Dr. Warren-Boulton’s opinion

that “[t]he relevant issue is whether the parties can arrive at a consistent understanding of the fair shares for each Drug at Issue at some point reasonably soon after the initial price increase, however different that may be across the Drugs at Issue.” Id. at 34 (noting that “[f]or Drug A [fair share] may be 50-50, while for Drug B that may be 60-40 or 80-20”). These opposing expert views underscore that the record here does not warrant granting summary judgment. When considered together, the circumstantial evidence presented—the phone records of communications between competitors contemporaneous with price changes, adverse inferences from Aigner, Nesta, and Aprahamian’s Fifth Amendment invocations, and stable market shares following a large price increase—“tends to exclude the possibility” that Mylan “acted independently” in deciding to follow Taro’s Phenytoin price increase or in allocating the Phenytoin market. Matsushita Elec. Indus. Co., 475 U.S. at 588. Because the States have raised a genuine dispute of fact as to whether Mylan joined the alleged Phenytoin conspiracy, Mylan’s motion for summary judgment on the States’ Phenytoin claims is DENIED.

IV. CONCLUSION Because there is evidence from which a reasonable jury could find that Mylan participated in conspiracies related to Bromocriptine, Phenytoin, and Pio Met, as well as the overarching conspiracy, the Motion for Summary Judgment is denied.

IT IS SO ORDERED.

/s/ Michael P. Shea, U.S.D.J.

Dated: Hartford, Connecticut January 9, 2026