From Casetext: Smarter Legal Research

In re HIV Antitrust Litig.

United States District Court, Northern District of California
Nov 1, 2023
19-cv-02573-EMC (N.D. Cal. Nov. 1, 2023)

Opinion

19-cv-02573-EMC

11-01-2023

IN RE HIV ANTITRUST LITIGATION.


ORDER DENYING PLAINTIFFS' MOTION FOR A NEW TRIAL

DOCKET NO. 2088

EDWARD M. CHEN UNITED STATES DISTRICT JUDGE

Currently pending before the Court is Plaintiffs' motion for a new trial. Plaintiffs argue that the clear weight of the evidence does not support the jury's responses to Questions (1) and (2), which concerned, respectively, market power in the relevant market and pay-for-delay. Having considered the parties' briefs, the evidence presented at trial, and the oral argument of counsel, the Court hereby DENIES Plaintiffs' motion.

I. FACTUAL & PROCEDURAL BACKGROUND

In the jury verdict form, there were a set of questions related to anticompetitive conduct and antitrust injury (causation and damages). There were four questions on anticompetitive conduct. Only the first two are relevant for purposes of the pending motion. Question (1) asked: “At step one of the rule of reason, did the plaintiffs prove that Gilead had market power within the relevant market(s) that included Truvada and/or Atripla?” Question (2) asked: “At step one of the rule of reason, did the plaintiffs prove that the April 25, 2014, patent settlement agreement between Gilead and Teva included a ‘reverse payment' from Gilead to Teva so that Teva would delay its entry into the market and Gilead could thereby avoid the risk of generic competition?”

On June 30, 2023, the jury indicated to the Court that it had reached a verdict. See Trial Tr. (Vol. 18), at 3406. However, before the jury gave its verdict to the Court, it posed the following question to the Court: “‘Do we have to answer Question Number 1? Can it be skipped if we have an answer for Question Number 2?'” Id. at 3406.

The Court conferred with the parties about the jury's question. The following exchange took place:

THE COURT: And I think the answer is: Yes. You can't skip anything. I propose just to say.
[DEFENSE]: We agree, Your Honor.
THE COURT: - yes, you have to answer Number 1. [¶] Okay. If there's no objection, I will just write on here: Yes, you must answer Question Number 1. [¶] Is that all right? [¶] Okay. [¶] Well, why don't we recess a little bit.
Id. at 3406-07.

Several minutes after the Court's response was delivered to the jury, the jury informed the Court that it was prepared to deliver its verdict. See id. at 3407. Below is the relevant portion of the verdict form. As reflected below, the jury answered both Questions (1) and (2), in spite of instructions on the verdict form indicating that a “no” answer on (1) would make any further responses unnecessary.

Element (1): Anticompetitive Conduct
1. At step one of the rule of reason, did the plaintiffs prove that Gilead had market power within the relevant market(s) that included Truvada and/or Atripla?
a. Truvada Yes ____ No √.
b. Atripla Yes _____No √.
If you answered "yesto either of the above, proceed to Question 2. If you answered "no ” to both of the above, skip the remaining questions, and sign and date the form.
2. At step one of the rule of reason, did the plaintiffs prove that the April 25, 2014, patent settlement agreement between Gilead and Teva included a “reverse payment” from Gilead to Teva so that Teva would delay its entry into the market and Gilead could thereby avoid the risk of generic competition?
Yes____ No √.
If you answered "yes" to the above, proceed to Question 3. If you answered "no, ” skip the remaining questions, and sign and date the form.

Docket No. 2057 (Verdict at 2). The responses to Questions (1) and (2) resulted in a verdict in favor of Defendants.

II. DISCUSSION

A. Legal Standard

Federal Rule of Civil Procedure 59 provides in relevant part as follows: “The court may, on motion, grant a new trial on all or some of the issues - and to any part - as follows: (A) after a jury trial, for any reason for which a new trial has heretofore been granted in an action at law in federal court.” Fed.R.Civ.P. 59(a)(1)(A).

“[A] district court can grant a new trial under Rule 59 on any ground necessary to prevent a miscarriage of justice.” Experience Hendrix L.L.C. v. Hendrixlicensing.com Ltd, 762 F.3d 829, 845-46 (9th Cir. 2014). For instance, “a trial court may grant a new trial if the verdict is contrary to the clear weight of the evidence” - “even if substantial evidence supports the jury's verdict.” Kode v. Carlson, 596 F.3d 608, 613 (9th Cir. 2010). But “‘[a] district court may not grant a new trial simply because it would have arrived at a different verdict.'” Wallace v. City of San Diego, 479 F.3d 616, 630 (9th Cir. 2007).

“[I]n considering [a] motion for a new trial, [a court is not] required to draw all inferences in favor of the verdict and [can] reweigh the evidence and make credibility determinations.” Experience, 762 F.3d at 845.

B. Responses to Both Jury Verdict Questions (1) and (2)

As a preliminary matter, the Court notes that the jury answered both Questions (1) and (2). Plaintiffs take the position that the Court should consider the response to Question (1) only and disregard the response to Question (2) as surplusage. Defendants take the position that the Court should consider the jury's responses on both questions.

Of course, Plaintiffs still argue in their motion that the jury erred on both questions, not just Question (1).

1. Waiver

Defendants make a strong case that, by failing to object before the jury was discharged, Plaintiffs have waived the right to argue that the Court should consider the response to Question (1) alone. Defendants point out that there multiple opportunities where Plaintiffs could have objected, or at least voiced a concern:

• when the jury first asked if it had to answer Question (1) since it had an answer for Question (2);
• when the Courtroom deputy read the jury's responses to both questions, see Trial Tr. (Vol. 18), at 3409 (Court prompting Courtroom Deputy to read the response to Question (2) for “completeness”); and
• after the jury was generally polled, see id. at 3409-10; and
• after the Court stated that it would discharge the jury “unless there's anything else that counsel wants me to raise or ask.” Id. at 3411.

Post-hearing, Plaintiffs submitted a letter in which they argue, inter alia, that it was not clear the jury had an answer for Question (2) because the jury asked if Question (1) could be skipped if it had an answer for Question (2). Plaintiffs ignore the context of the jury's inquiryi.e., this question came from the jury after it initially informed the Court that it had reached a verdict. Thus, implicitly but unmistakably, the jury had reached a consensus on Question (2).

Plaintiff's primary response is that objecting would have been a pointless exercise because “there was nothing for the jury to rectify....Had the plaintiffs raised the issue before the jury was dismissed, they would have requested the exact same relief sought here: that the Court set aside the jury's answer to Question 2 as surplusage.” Reply at 10. The Court is not convinced.

Plaintiffs could have, e.g., asked that the jury be directed to follow the instructions on the verdict form, see Floyd v. Laws, 929 F.2d 1390, 1396 (9th Cir. 1991) (indicating that a trial court has the authority to “resubmit[] [a] special verdict to the jury, provided the jurors have not been discharged”), and it is far from clear what the jury would have done. Had the jury been instructed to answer Question (1) and not gone on to Question (2) if it answered “no” on (1), and if it had devoted most of its deliberations on (2) and less time on (1), the jury could have extended its deliberations on (1) and possibly have reached a different verdict. The jury could also have asked for clarifying instructions on the verdict form because its responses to Questions (1) and (2) in favor of Defendants were consistent and there was no logical reason why it could not answer both - particularly because the issues in (1) and (2) (market power in a relevant market and pay-for-delay, respectively) were both part of the first step of the rule-of-reason inquiry, such that a “no” response on either (1) or (2) would result in no liability to Defendants.

See Docket No. 2018 (Jury Instruction No. 12) (“The rule of reason has a three-step, burdenshifting framework. [¶] First, the plaintiffs have the initial burden of showing that the defendants' conduct produced significant competitive effects (as defined in later Instructions No[s]. 13-16) within a relevant market.”) (emphasis in original); Docket No. 2018 (Jury Instruction No. 13) (“As I noted, at the first step of the rule of reason, the plaintiffs have the initial burden of showing that the defendants' conduct produced significant anticompetitive effects within a relevant market. According to the plaintiffs, the defendants' conduct produced significant anticompetitive effects when, as alleged, Gilead made to Teva what is sometimes called a ‘reverse payment' ....[¶] For a reverse payment to have an anticompetitive effect, the plaintiffs have the burden of proving the following: (1) Gilead had market power; and (2) The patent settlement agreement . . . included a payment from Gilead to Teva so that Teva would delay its entry into the market and Gilead could thereby avoid the risk of generic competition.”).

Post-hearing, Plaintiffs suggested for the first time that waiver has no place in a situation such as the one before the Court - i.e., it is generally permissible to raise an issue with the verdict after the jury has been discharged. The cases Plaintiffs cite, however, are distinguishable. For instance, in Tritchler v. County of Lake, No. 98-16062, 2000 U.S. App. LEXIS 17463 (9th Cir. July 17, 2000), the Ninth Circuit did reject the plaintiffs' contention that the defendants had “waived their objections to inconsistent verdicts by failing to insist that the verdicts be resubmitted, instead suggesting that the issue be resolved by way of post-trial motion.” Id. at *4.

However, the court explained that this was because

[t]he issue was raised before the jury was excused, cf. Home Indem. Co. v. Lane Powell Moss & Miller, 43 F.3d 1322, 1331 (9th Cir. 1995) (holding counsel risks waiver by failing to raise the issue before the jury is excused then counsel risks waiver), and counsel did not affirmatively mislead the court, see Los Angeles Nut House v. Holiday Hardware Corp., 825 F.2d 1351, 1354-55 (9th Cir. 1987) (party does not waive objections by failing to object immediately and move for resubmission absent counsel affirmatively misleading the court).
Id. (emphasis added). The basis for rejecting waiver in Tritchler does not apply here.

As for Los Angeles Nut House v. Holiday Hardware Corp., 825 F.2d 1351 (9th Cir. 1987), there, the Ninth Circuit did reject the argument that, “when a jury's answers to written interrogatories are inconsistent with its general verdict, under Fed.R.Civ.P. 49(b) a party waives objection to the inconsistency by failing to object immediately and move for resubmission of the inconsistency before the jury is dismissed.” Id. at 1354. However, the case is not on point because it addressed the situation where there is an inconsistency between a general verdict and answers to written questions, in which case Federal Rule of Civil Procedure 49(b) specifies how the trial court is to deal with this situation. See id. (“[Rule 49(b)] gives a district judge three options for resolving any inconsistency between the verdict and answers to special interrogatories [i.e., judgment may be entered in accordance with the answers, the jury may be directed to further consider its answers and verdict, or a new trial may be ordered].... We will not amend the rule and provide a fourth alternative, one that permits the party in whose favor the general verdict runs to prevail merely by remaining silent.”). It is also worth noting that Rule 49(b)(2) provides: “When the general verdict and the answers are consistent, the court must approve, for entry under Rule 58, an appropriate judgment on the verdict and answers.” Fed.R.Civ.P. 49(b)(2) (emphasis added). Los Angeles Nut House's rejection of waiver under Rule 49(b) does not apply here.

Finally, in Pierce v. Southern Pacific Transportation Co., 823 F.2d 1366 (9th Cir. 1987), the Ninth Circuit did state that “[s]pecial verdicts . . . are governed by Rule 49(a), which does not require objections before discharge of the jury.” Id. at 1370. The instant case, however, involves a general verdict with answers to written questions (Rule 49(b)) and not a special verdict (Rule 49(a)). See, e.g., 9 Moore's Fed. Prac. - Civ. § 49.02[2][b] (explaining that, for a special verdict, the jury provides written answers to questions concerning specific factual issues and “[t]he trial court then applies the law to those answers and enters judgment accordingly[;] [b]y removing from the jury the consideration or application of the law (which a general verdict requires through application of the court's instruction), the special verdict avoids those two sources of possible error in the general verdict”); Embroidery Indus. v. Brasking, Inc., 60 Fed.Appx. 111, 114-15 (9th Cir. 2003) (stating that “the critical difference between a special verdict and a general verdict is that, in a special verdict under Rule 49(a), the jury is requested to make findings limited to specific factual issues, while in a general verdict pursuant to Rule 49(b), the jury must respond to special interrogatories concerning its factual findings, and, in addition, decide which party should prevail after applying the law to the facts”); Floyd, 929 F.2d at 1395 (noting that “[j]uries rendering general verdicts face a dual task” - i.e., “they are responsible for finding facts” and “they must reach a general verdict by applying the law to their findings of fact[;] [t]o help the jury reach a general verdict, the court must often provide a very long and detailed explanation of the law”).And even if that were not the case, Pierce's concern was about potential inconsistent responses made by the jury, see Pierce, 823 F.2d at 1370 (stating that a “court has a duty to reconcile the jury's special verdict responses on any reasonable theory consistent with the evidence”), and, here, there were no inconsistent responses. Where the responses are inconsistent and cannot be reconciliated, resubmission to the jury under Rule 49(a) prior to their discharge is permissible. Floyd, 929 F.2d at 1396; yet Plaintiffs failed to request resubmission.

Cf. Wright & Miller, Fed. Prac. & Proc. § 2501 (“To classify what type of verdict is rendered, the totality of the jury instructions must be examined; if it is determined that the jury made a finding on the ultimate question of liability, then it is classified as a general verdict. This classification can determine whether or not the right to appeal an inconsistent jury verdict is waived.”).

2. Surplusage

In any event, even if Plaintiffs did not waive their right to challenge the response to Question (2), the Court still finds it appropriate to consider the response. Plaintiffs' argument that the Court should disregard the response to Question (2) as surplusage is based on the Ninth Circuit's decision in Floyd.

In Floyd, the plaintiff filed a § 1983 suit against a city and its police of chief. At trial, the jury returned a special verdict that was ultimately favorable to the defendants. After the jury was discharged, defendants pointed out that the jury's answers to Questions 13 and 14 appeared inconsistent. Question 13 asked whether the plaintiff was damaged as a result of the police chief's actions, and the jury answered “no.”

After question 13, appeared the following instruction: "If your answer to question 13 is 'No,' do not answer any further questions, but proceed to the end of this form and sign the verdict. If you answered 'Yes' to question 13, proceed to question 14." Question 14 read: "What amount of money will reasonably compensate plaintiff Debbie Floyd for any of the actions of defendant Laws' found in questions 9, 10, 11, or 12?" To which the jury responded: "$ 7,500.00."
Id. at 1393 (emphasis added). The trial court found that the jury's response to Question 14 was “surplusage because, after replying ‘No' to question 13, the jury disobeyed the express instructions of the verdict form and answered question 14.” Id.

The Ninth Circuit began by noting that, in Gallick v. Baltimore & O.R.R. Co., 372 U.S. 108 (1963), the Supreme Court held that, “when confronted with seemingly inconsistent answers to the interrogatories of a special verdict, a court has a duty under the seventh amendment to harmonize those answers, if such as possible under a fair reading of them.” Floyd, 929 F.2d at 1396. “[O]nly if all attempts at reconciliation fail” may a court order a new trial. Id. However, the Ninth Circuit then noted that Gallick did not “address the situation in which a jury proffers superfluous answers in violation of a trial court's express instructions contained on the special verdict form.” Id. at 1397. The Ninth Circuit held that “special findings issued in violation of the trial court's express instructions do not constitute legitimate or viable findings of fact. The trial court must therefore dismiss them as surplusage, as a matter of law.” Id. (emphasis added). In support of its holding, the Ninth Circuit cited White v. Grinfas, 809 F.2d 1157, 1161 (5th Cir. 1987), and Tanno v. S.S. Pres. Madison Ves, 830 F.2d 991, 993 (9th Cir. 1987). Importantly, in each case, as in Floyd, there was a contradiction in the jury's verdict. See Floyd, 929 F.2d at 1397 (noting that, in Tanno, the jury's verdict contained parenthetical comments that “exposed an implicit contradiction”; the parenthetical comments indicated that the jury calculated pain and suffering for six days, but “[a]ll the evidence presented . . . indicated that Tanno had suffered pain for a far longer period than six days”); Id. at 1398 (noting that, in White, the jury's response to one question indicated that the counterdefendants withheld information about structural defects with intent to defraud, but the response to another question stated that they had no such intent). The existence of the inconsistency framed the issue in Floyd and the cases it cites.

The basic question for the Court is whether Floyd forecloses an argument that the surplusage rule applies only when the jury's verdict contains inconsistent responses, and not where (as here) there are consistent responses. That was the case in Floyd as well in the two cases on which it relied. Plaintiffs contend that the argument is foreclosed because the Ninth Circuit expressly identified the problem as the jury failing to follow instructions; if the failure to follow instructions is the key, then it does not matter whether a jury's responses are consistent or inconsistent.

The Court rejects Plaintiffs' position for several reasons. First, Floyd was clearly concerned with inconsistencies in a jury verdict, as indicated by the header “APPARENT INCONSISTENCIES IN JURY'S SPECIAL VERDICT.” Id. at 1394. Notably, the Ninth Circuit's discussion of surplusage (in subsection (C)) came beneath this header. The inconsistency in the verdict answers provided the context and impetus of the court's decision. Where there is an inconsistency, the court has a duty under the Seventh Amendment to try to reconciliate the responses. Floyd, 929 F.2d at 1396. If that is impossible, the court can, prior to discharge of the jury, resubmit as an option. See id. It may also order a new trial. Id. Another option is to disregard the inconsistent response if that response can be deemed superfluous. See id. at 1397-1401. The critical predicate in Floyd was an inconsistent verdict, a predicate not present in the case at bar. Thus, there is no principled reason why the surplusage rule should apply where there are consistent responses in a jury verdict. In fact, what the jury did here is similar to what courts often do in an addressing a case: offer independent reasons supporting its ruling.

Second, the Court does not agree with Plaintiffs' contention that the White case, on which the Ninth Circuit relied in Floyd, demonstrates that the Ninth Circuit was simply concerned about a jury's failure to follow instructions, i.e., regardless of whether the jury's responses were consistent or inconsistent. As indicated above, the text in Floyd demonstrates otherwise. See, e.g., Id. at 1396 (discussing Gallick where the Supreme Court was “confronted by seemingly inconsistent answers to the interrogatories of a special verdict”). Furthermore, in White, the Fifth Circuit made repeated note of “conflict[ing]” jury responses, thus suggesting that this fact was significant. White, 809 F.2d at 1161. That the Fifth Circuit stated that the conflicting responses were “caused by the jury's failure to follow the court's instructions,” and that responses that violated the instructions were “irrelevant,” Id., does not read out of the court's analysis the factual predicate that there were inconsistent jury responses. See also id. (“To effectuate best the intent of the jury, we hold that if the district court has correctly found that the jury's answer to a question that was supposed to determine further inquiry is clear and disposes of the legal issues, on review we must ignore the jury's necessarily conflicting answers to any other questions.”).

Third, there is a Ninth Circuit case that lends support to Defendants' position, even if it is not dispositive. Specifically, in Haynes v. Home Depot USA, Inc., No. 21-55827, 2023 U.S. App. LEXIS 836 (9th Cir. Jan. 13, 2023), the Ninth Circuit noted as follows:

The district court did not err by accepting an inconsistent verdict. The jury found that 1) Haynes did have ‘notice or information of circumstances to put a reasonable person on inquiry to seek to learn the facts necessary to bring an age discrimination claim by February 12, 2014,' and that 2) Haynes did not ‘prove that his age was a substantial motivating reason for Home Depot's decision to terminate him.' Although the jury ignored the instruction not to proceed to the second question based on its answer to the first, these answers are consistent and provide no basis for reversal.
Id. at *6. Plaintiffs argue that Haynes is “inapposite because the only question before the court was whether the plaintiff was entitled to a new trial based on the district court's acceptance of an inconsistent verdict....[T]he court had no reason to consider the applicability of the surplusage rule because the jury's response to the first question was dispositive and unchallenged.” Reply at 9-10. Plaintiffs are correct that the Haynes court did not expressly refer to the surplusage rule. Nevertheless, Haynes does suggest that, in situations where jury responses are consistent, there are not the same concerns which drove the court in dealing with inconsistent answers in Floyd.

Finally, it is worth noting that, in the case at bar, there is a fair argument that the jurors were not ignoring instructions but rather may have found some tension within the instructions (to the extent they were instructed to answer Question 1 first). As Defendants assert:

[I]t was entirely reasonable for the jury to start with Question 2. [The jurors] were instructed [in the jury instructions and the verdict form instructions] that market power and payment-for-delay are both part of the “first step” of the rule of reason, so a “no” verdict on Question 2 would preclude liability regardless of Question 1.

Opp'n at 12 (emphasis in original); cf. Sicilia v. Safeco Ins. Co. of Am., No. 89-56096, 1992 U.S. App. LEXIS 17761, at *9-10 (9th Cir. July 23, 1992) (in insurance case, upholding jury verdict even though jury did not follow instructions; jury perceived correctly that there was a flaw in the instructions and thus validly responded that plaintiff could be awarded damages for one fire but not for a second fire: “the verdict was not at all inconsistent, and was easily reconcilable”).

Taking into account all of the circumstances above, the Court concludes that it is appropriate in this case to consider the jury's responses to both Questions (1) and (2). In now turns to each of those questions.

C. Jury Verdict Question (1): Market Power within the Relevant Market

Plaintiffs argue that there is a miscarriage of justice that warrants a new trial because the clear weight of the evidence goes against the jury's response to Question (1) which concerns market power within the relevant market. Plaintiffs address the issue of market power first and then the issue of relevant market, but that approach is out of order - i.e., one needs to have a sense of the relevant market before evaluating market power.

1. Relevant Market

The jury instruction on relevant market was Instruction No. 15, which provided in part as follows:

[A] relevant product market is defined as the product at issue and economic substitutes for that product. A product is an economic substitute for the product at issue if the sales of the other product could substantially constrain the ability of the manufacturer of the product at issue to increase the price of the product. In making that determination you should consider whether the two products are reasonably interchangeable in terms of use or whether consumers will change their consumption of one product in response to a price change in another.
For example, you may consider whether a small but significant and
non-transitory increase in the price of one product, from the competitive level, would result in enough customers switching from that product to another product such that the price increase would not be profitable. In other words, would customers accept the price increase or would so many switch to alternative products that the price increase would be withdrawn? Generally speaking, a small but significant and non-transitory increase in price is approximately a 5 percent increase in price not due to cost factors but you may conclude in this case that some other percentage is more applicable to the product at issue. If you find that customers would switch and that the price increase would not be profitable, then you may conclude that the other products are in the product market. If, on the other hand, you find that customers would not switch, and the price increase would be profitable, then you may conclude that the products are not in the product market.
In evaluating whether various products are economic substitutes for each other, you may also consider:
• Consumers' reviews on whether the products are interchangeable;
• the relationship between the price of one product and the sales of another;
....
• the perceptions of either industry or the public as to whether the products are in separate markets ....
Docket No. 2018 (Jury Instruction No. 15, at 21-22). Accordingly, the basic question for the jury was what other drugs constrained the price of Truvada and Atripla? Plaintiffs argue that the clear weight of the evidence supports only generic versions of the drugs being such constraints. Plaintiffs contend that only generics stopped the “‘relentless[]'” increase of prices - “more than 6% per year for many years.” Mot. at 11; see also Mot. at 6 (noting that “inflation [was] at only 2%).

Plaintiffs argue that Defendants improperly argued to the jury that reasonable interchangeability of use was all that was needed for products to be in the same market. The Court does not agree. Although Defendants emphasized this factor to the jury, they did not contend that the jury could thereby ignore whether the interchangeable products substantially constrained Gilead's ability to increase the prices of Truvada and Atripla.

The Court cannot say that the clear weight of the evidence supports Plaintiffs' theory that the brand drug and its generic version were the only two drugs in the relevant market. Essentially, there were competing experts (Dr. McGuire for Plaintiffs and Dr. Wu for Defendants), and several of Dr. Wu's criticisms of Dr. McGuire's opinions were fair - especially since, as Defendants point out, (1) Dr. McGuire did not provide much testimony on the issue of relevant market and (2) Dr. McGuire never came back as a witness to rebut Dr. Wu's criticisms.

For example, it is clear that other drugs were available to treat HIV, and, implicitly, some of those drugs acted as a constraint on what Gilead could do with Truvada and Atripla in terms of price because, as Dr. Wu explained, Gilead's drugs “aged” and newer, more innovative treatments came onto the market. See Tr. (Vol. 13), at 2668 (Wu testifying that, by the time of generic launch in 2020, Atripla and Truvada were no longer the gold standard - i.e., there were not on DHHS's list of preferred options); Tr. (Vol. 13), at 2672 (Wu testifying that, in 2015, Atripla was now indicated as an alternative first-line therapy, not a preferred therapy; adding that, in 2022, the only preferred single-tablet regimens were Triumeq, Biktarvy, and Dovato). There was also evidence that consumers did switch to newer, more innovative drugs - including non-Gilead drugs - prior to generic entry. See, e.g., Tr. (Vol. 13), at 2659-62 (Wu testifying that Patient A started with Truvada/Reyataz/Norvir (multitablet); then switched to Atripla (single tablet) in 2013; then switched to Triumeq (single tablet) in 2016; then switched to Descovy/Tivcay (two pills) in 2018; and then switched to Biktarvy in 2020; patient did not thereafter switch to a generic). And notably, for Atripla, sales had dropped significantly (90%) before generic entry. Thus, newer, more innovative drugs arguably placed some constraint on Gilead's ability to price.

That Gilead was still able to increase the price of its drugs by 6% every year is, in and of itself, not dispositive. First, as indicated above, even though Gilead was increasing the price of its drugs, the sales of those drugs were falling before generic entry. Second, there was some evidence that Gilead was aware its ability to price was affected by other nongeneric drugs. See Ex. 306 (slides on Gilead HIV Generic Impact Project, Summary of Senior Management Interviews) (at slide 4, noting that one Gilead employee suggested, back in September 2008, “What we can do” is “[i]ncrease price as much as possible while we still can” before generic entry but “may be too late” to “differentiate[] FTC from 3TC”) (emphasis added).

Implicitly, Plaintiffs recognize that it is a harder case for them to win on Atripla (whose sales, as noted above, dropped by 90% before generic entry). Thus, in their opening brief in particular, they focus on Truvada. At trial, Dr. Wu testified that, prior to generic entry, Truvada sales had also dropped by 90%. But Plaintiffs contend (as they did at trial) that this figure was deceptive: there was a drop of 90% only if sales of Truvada for purposes of treating HIV were considered. Dr. Wu ignored sales of Truvada for purposes of PrEP, even though (1) as of 2020, about 90% of all sales of Truvada were for PrEP, see Tr. (Vol. 13), at 2702 (Wu admitting that “[t]hat sounds about right to me”), and (2) during the relevant time, Truvada was the only drug approved for PrEP up until Descovy, which was just another Gilead drug and thus (according to Plaintiffs) could not be a real constraint on the pricing of Truvada.

Truvada arguably presents a closer call; nevertheless, the clear weight of the evidence does not run in Plaintiffs' favor. Plaintiffs never claimed as a relevant market the market for drugs used for PrEP. And even if the relevant market claimed was Truvada or its generic for all of its uses, Plaintiffs did not give a clear explanation as to why all of its uses was the correct benchmark. Furthermore, as Dr. Wu testified, even if all Truvada sales were counted (both for treatment and for PrEP), there was still a decline in Truvada sales prior to generic entry, around 50%. See Tr. (Vol. 13), at 2708. At the hearing, Defendants underscored that there was this decrease even before Descovy came onto the market; Plaintiffs did not dispute that contention.

Accordingly, the Court concludes that the clear weight of the evidence does not support Plaintiffs' asserted relevant markets (i.e., a market consists of only the brand drug and its generic version). There was no miscarriage of justice therefore for the jury to answer “no” in response to Question (1).

2. Market Power

Because the clear weight of the evidence does not support Plaintiffs' claimed relevant markets, the Court need not consider the issue of market power. However, even if the Court were to consider the matter, it still would not rule in Plaintiffs' favor. According to Plaintiffs, the clear weight of the evidence establishes that Gilead has market power. They point to: (1) the charging of supracompetitive prices until generic entry; (2) Gilead's high gross margins; and (3) Gilead's reduction of output.

With respect to (1), the charging of supracompetitive prices, Plaintiffs' position largely replicates their position on the issue of relevant market - i.e., the competitive price is the generic price because the market consists of the brand and the generic. Similarly, for (3), Gilead's reduction of output, Plaintiffs' argument is tied to the product market being limited to the brand and its generic version. See Mot. at 9 (arguing that “Truvada output increased after generic entry” and that insurers were not able “to obtain discounts or rebates on Truvada and Atripla”).

With respect to (2), Gilead's high gross margins, Plaintiffs fairly point out that the gross margins were high. But similar to above, there were competing experts on, e.g., whether R&D costs factored into the price of Gilead's drugs. Dr. McGuire asserted they did not. See Tr. (Vol. 9), at 1646 (“High R&D spending does not cause high prices for patent products; it's the other way around. The ability to set high prices after you've got a patent that protects you from competition is what motivates you to engage in the R&D. [¶] ....[S]uppose someone just gave Gilead these patents for free, they fell out of the sky, ....[¶] Would anyone think that Gilead would be charging anything different than $51 for Truvada and $84 for Atripla prior to generic entry?”). His position was a fair one. However, that does not mean that Dr. Wu's contrary position was unreasonable. In fact, the jury reasonably could have thought that R&D costs did factor into pricing and that the possibility of getting a patent motivated Gilead to do R&D.

To the extent Plaintiffs argue that Gilead's gross margins on Truvada and Atripla exceeded those of pharmaceutical companies generally (97-98% compared to 77%), that is true. However, on cross, Dr. McGuire admitted that he “wouldn't be surprised” if some of the pharmaceutical companies in the study at issue also reported gross margins over 90%. Tr. (Vol. 9), at 1719. Furthermore, Dr. McGuire admitted that the 77% figure he cited was based on “public filings [of the pharmaceutical companies that] don't include drug-specific information.” Tr. (Vol. 9), at 1720. If one were to consider non-drug-specific information for Gilead, Gilead's public filings reported gross margins in a similar range: 78%. See Tr. (Vol. 9), at 1720. Moreover, there was evidence that Gilead's gross margins for Atripla specifically were less than 70%. See Tr. (Vol. 9), at 1721-22 (McGuire conceding such). Dr. McGuire concluded that gross margins were 97% because he believed a correction needed to be made “to portray the revenue and costs with respect to Gilead alone as opposed to Bristol Meyers Squibb” who contributed one of the component drugs to Atripla. Tr. (Vol. 9, at 1724); see also Mot. at 8 n.1 (arguing that it is “logical” that “Gilead's gross margins on Atripla and Truvada were the same . . . given that Atripla is comprised of Truvada plus BMS's efavirenz, and Gilead turns over the portion of the Atripla revenues associated with efavirenz to BMS”). However, that point was reasonably disputed.

But see Mot. at 8 (arguing that the “77% pharmaceutical industry median is misleading: these margins are also high because many of these companies ‘sell branded products protected by patents' and thus likely have market power with respect to such products”) (emphasis in original).

In their reply, Plaintiffs argue that “[l]ogic and the law dictate that margins are appropriately measured with respect to the drugs at issue, not Gilead's entire drug portfolio.” Reply at 3. This is a fair point. However, if Dr. McGuire was looking at other pharmaceutical companies' entire drug portfolio for the 77% figure, it was fair for Defendants to point to Gilead's gross margins on its entire drug portfolio.

D. Jury Question (2): Pay-for-Delay

As indicated above, Plaintiffs argue that the Court should not even consider the response to Question (2) on the basis that it was surplusage (i.e., because the jury violated the Court's instructions). The Court has rejected that contention and thus considers the jury's response to Question (2). It does so out of an abundance of caution; Plaintiffs' motion for a new trial may be denied based on their failure to show a miscarriage of justice on Question (1) alone.

For Question (2), there are two major issues for the Court to take into account: (1) was there a reverse payment and (2) was it made to delay generic entry (e.g., was it large and unexplained)?

Plaintiffs presented a reasonable case that there as a large and unjustified reverse payment which would then allow the jury to infer that the payment had been made to delay generic competition. See Docket No. 2018 (Jury Instruction No. 16, at 23) (“If the brand manufacturer's payment to the generic manufacturer is large and unjustified, then you may infer that the brand and generic manufacturers agreed that the brand manufacturer would pay the generic manufacturer to delay the generic manufacturer's entry into the market and thereby avoid the risk of generic competition.”). However, that does not mean that the clear weight of the evidence favors Plaintiffs. See Opp'n at 14 (“[T]ellingly, Plaintiffs did not even file a Rule 50(a) motion.”).

First, even if, as Plaintiffs argue, the exclusivity provision and entry date were negotiated at or about the same time (contrary to Defendants' position that the entry date was decided before exclusivity was ever discussed), there was evidence that there was no large payment to Teva.

More specifically, there was evidence that, at the time of the settlement agreement, the exclusivity provision was not expected to have significant value. See Apotex, Inc. v. Cephalon, Inc., 255 F.Supp.3d 604, 611 (E.D. Pa. 2017) (stating that “he relevant rule of reason analysis is conducted on an ex ante basis, that is, as of the time the settlements were executed”). This is because there were other paths for a generic manufacturer to get into the market ahead of, or at the same time as, Teva.

In their papers, Plaintiffs argue that they did not have to prove that exclusivity would be guaranteed. See, e.g., Reply at 12. This is true; however, evidence that there were other paths support Defendants' position that any payment was expected to be of little or speculative value.

The FTC Settlement Agreement provided as follows:

“License Effective Date” means, with respect to a particular Teva ANDA Product or a particular FTC Generic Equivalent, on an FTC Combination Product-by-FTC Combination Product basis, the earliest to occur of the following dates:
(i) September 30, 2020;
(ii) the date on which a Final Court decision is entered holding that each of the ten-asserted unexpired patent claims included in the Licensed FTC Patents that were asserted against Teva in the Litigation is invalid or unenforceable;
(iii) the date on which all of the Licensed FTC Patents listed in the Orange Book, with respect to the applicable Gilead Product or FTC Product, has been permanently abandoned or delisted from the Orange Book (excluding any such Licensed FTC Patents as have, as of such date, been held to be invalid or unenforceable);
(iv) with respect to Teva ANDA Product only, and except with respect to any Access Use, to the extent that Gilead, any of its Affiliates, or a Third Party (under authorization from Gilead or its Affiliates, by license, sublicense, waiver, covenant not to sue or otherwise) plans or is authorized to sell the relevant FTC Generic Equivalent in the Territory on a date that falls any time before September 10, 2021, the date that is six (6) months prior to (1) the sale thereof by Gilead or its Affiliate or (2) the earliest permitted sale thereof by such Third Party, in the Territory; and
(v) With respect to Teva ANDA Product only, and except with respect to any Access Use, to the extent that Gilead, any of its Affiliates, or a Third Party (under authorization from Gilead or its Affiliates, by license, sublicense, waiver, covenant not to sue or otherwise) plans or is authorized to sell the relevant FTC Authorized Generic in the Territory on
a date that falls at any time before September 10, 2021, the date such FTC Authorized Generic is first sold in the Territory.
Ex. 2000 (FTC Sett. Agmt. § 1.1).

Item (iv) above is the clause that Plaintiffs challenged. Under the clause, Teva's entry date is accelerated so that it is six months (180 days) ahead of the entry date for another generic manufacturer. In other words, if Gilead were to give a license to another generic manufacturer to make generic Truvada or Atripla, then Teva's date of entry would be six months (180 days) ahead of any sale by the other generic manufacturer.

But there are other ways for a generic manufacturer to enter the market ahead of, or at the same time, as Teva. A generic manufacturer could get to the market ahead of Teva by launching at risk. A generic could get to the market the same time as Teva by initiating litigation and invalidating Gilead's patents (item (ii)). A generic could also get to the market the same time as Teva by getting Gilead's consent to sell an authorized generic (item (v)). To be sure, Dr. McGuire dismissed these other paths as unlikely. However, if, as Plaintiffs claimed, Gilead's patents were weak, then a generic manufacturer would have an incentive to launch at risk. See Tr. (Vol. 9), at 1755 (McGuire responding “[y]es” to the following question: “[I]f a party, one of the generics, had launched at risk and it really believed that [Gilead's] patents were weak, it could go to market before Teva, have an advantage before Teva, because in that scenario, a Court would not ultimately rule that [Gilead's] patents were valid,” and, “if that were true, an at-risk launcher has zero losses and an enormous gain; isn't that right?”). Similarly, if Gilead's patents were weak, then Gilead would have an incentive to give a generic manufacturer challenging its patents permission to sell an authorized generic. See Opp'n at 20 (arguing that, if a generic manufacturer posed a credible threat, then Gilead could simply allow an authorized generic; “how could the 180-day clause be worth so much if Gilead could simply override Teva's potential headstart?”).

In their reply, Plaintiffs argue that none of the other paths happened and that, instead, every other generic did not enter the market until 180 days after Teva. But that evidence could also be read to mean that Gilead's patents were strong, which goes against Plaintiffs' theory that there was a payment for delay.

Second, in the papers, Defendants underscore that, even if there were some kind of payment to Teva, there was no payment for delay (i.e., no quid pro quo) because there was evidence that Gilead's patents were strong and the strength of Gilead's patents explained the entry date of September 2020 that was provided for in the FTC Settlement Agreement. The Court agrees. At trial, Plaintiffs took the position that it was highly likely that Teva would prevail in the infringement suit that Gilead had brought with respect to the FTC patents, see Tr. (Vol. 7), at 1286 (Mr. Lentz testifying that there was a 75-80% chance that Teva would have prevailed in the infringement suit), and that it was highly likely that Gilead's follow-on patents would be invalidated. See Tr. (Vol. 15), at 3214 (Plaintiffs arguing in closing that “Mr. Lentz concluded that Teva had a 75 to 80 percent chance of prevailing and invaliding Gilead's weak FTC patents in the FTC litigation, and an 80 to 90 percent of invaliding the even weaker combination patents”); Docket No. 1407-5 (Lentz Expert Rpt. ¶¶ 278, 363) (opining that Teva's chance of success with respect to the FTC patents was around 75-80% and with respect to the follow-on patents around 90%); Tr. (Vol. 9), at 1656 (Dr. McGuire essentially agreeing that he was instructed by Plaintiffs' counsel to assume that “Teva's combined chance of winning against all of the relevant patents, [i.e.,] the FTC patents and the combination patents, was 62.6 percent”). Taking into account both of those considerations, Plaintiffs took the position that generic entry would have taken place in May 2019 if Defendants had not engaged in anticompetitive conduct. See Tr. (Vol. 9), at 1635 (McGuire discussing different entry dates dependent on different probabilities of defeating Gilead's patents).

But if the jury disagreed with the high level of probabilities claimed (which it reasonably could have done), then the entry date would have been later than May 2019. For example, if the jury believed Ms. Julie's testimony that Teva's chances in the patent litigation were 50% and that its chances on the follow-on patents were 75%, then the combined likelihood of success would have been 37.5%, which was not even on the table in Dr. McGuire's report. See Tr. (Vol. 9), at 1635 (Wu testifying that a 50% chance of success yielded an entry date of February 2020); see also Docket No. 1376-5 (McGuire Rpt. ¶ 204 & Table 3) (“The Timing of Teva Generic Entry in Alternative Settlements Based on Teva's Chance of Winning the Truvada/Atripla FTC Patent Litigation”). Moreover, it was far from clear that even percentages at this level were justified. For instance, the defense patent law expert, Ms. O'Malley, provided persuasive testimony that Teva's likelihood of success on the FTC patents was low given that, e.g., no enantiomer patents had ever been invalidated. See Trial Tr. (Vol. 13), at 2574-75; see also Trial Tr. (Vol. 12), at 2517-19 (testifying about separation of negative and positive enantiomers). The defense scientific experts also provided persuasive testimony in support of Defendants' position that the FTC and combination patents were valid, which thereby impacted Teva's likelihood of success. See, e.g., Trial Tr. (Vol. 12), at 2421 (Dr. Chyall testifying, with respect to FTC patents, that glove analogy made by Plaintiffs' expert Mr. Lentz (an attorney) regarding separation of enantiomers was oversimplified); Trial Tr. (Vol. 11), at 2264-65 (Dr. Berkland testifying, with respect to the combination patents, why three approved drugs could not just simply be combined into one pill -e.g., “every time you add another active ingredient, you now have potential for stability problems with the new active ingredient”; making an analogy to combining Mentos and Coke).

Notably, in their opening motion, Plaintiffs failed to address as a substantive matter why the clear weight of the evidence went in their favor on the patent merits. In their reply brief, Plaintiffs protest that the “patent merits are relevant only to causation - whether, even absent the payment, Teva would not have been able to enter earlier due to valid and infringed patents.” Reply at 15 (emphasis added). Plaintiffs add that a large and unexplained reverse payment “obviates any need for the jury to even reach the patent merits.” Reply at 15. But Plaintiffs are wrong on both points. The patent merits informs the likelihood that a generic manufacturer would launch at risk, which was a way that a generic could get to market ahead of Teva in spite of the contractual exclusivity provision. Also, even though a large and unexplained reverse payment allows a jury to infer pay-for-delay, that does not mean that a defendant is barred from arguing no pay-for-delay because the patent owned by defendant was strong. Defendants made such an argument here and supported their position with substantial evidence.

In short, the clear weight of the evidence does not favor Plaintiffs on Question (2). The jury reasonably could have found that the contractual exclusivity provision was of limited or speculative value and/or that the entry date was not delayed based on a pay-for-delay (i.e., a quid- pro-quo) but rather was a fair reflection of the patent merits.

III. CONCLUSION

For the foregoing reasons, the Court denies Plaintiffs' motion for a new trial. Plaintiffs have not shown a miscarriage of justice, either with respect to the issue of market power in a relevant market or the issue of pay-for-delay. Plaintiffs rolled the dice by taking extreme positions - positions that the jury rejected. The Court now deems Phase I of the trial complete.

The Court sets a status conference for January 9, 2023, at 2:30 p.m., so that it may discuss with the parties how Phase II of the trial should proceed. A joint status report shall be filed one week prior to the conference.

This order disposes of Docket No. 2088.

IT IS SO ORDERED.


Summaries of

In re HIV Antitrust Litig.

United States District Court, Northern District of California
Nov 1, 2023
19-cv-02573-EMC (N.D. Cal. Nov. 1, 2023)
Case details for

In re HIV Antitrust Litig.

Case Details

Full title:IN RE HIV ANTITRUST LITIGATION.

Court:United States District Court, Northern District of California

Date published: Nov 1, 2023

Citations

19-cv-02573-EMC (N.D. Cal. Nov. 1, 2023)