court upholds nine-figure verdict in false advertising case

Guardant Health, Inc. v. Natera,Inc., 2025 WL 2106522, No. 21-cv-04062-EMC (N.D. Cal. Jul. 28, 2025)

Previous opinion discussed here.Guardant sued Natera for falsely advertising a clinical test; after the courtdenied a PI, it conducted a trial at which Natera was held liable for willful falseadvertising under the Lanham Act and California law. The jury awarded $75million in damages, awarded $175 million in punitive damages under California law,and issued an advisory verdict recommending $42 million in disgorgement. Italso rejected Natera’s false advertising counterclaims. The court here resolvespost-trial motions, including some discussion of the jury instructions.

The jury foundNatera’s advertisements were false by necessary implication. The ads comparedspecific performance metrics of Signatera and Reveal, the parties’ competingproducts, from two separate studies. “[M]issing from the side-by-sideadvertisement was adequate context of the differences in the studies that couldlead to further explanation why specific key metrics were not actuallycomparable on an apples-to-apples basis.” The comparisons, in context, were “statisticallyinvalid and misleading” because of differences in the underlying datasources. This wasn’t puffery. Natera’s advertisements presented “specific andmeasurable” claims of superiority of Signatera over Reveal.

Advertisements using an “ ‘apples-to-oranges’ theory offalsity” are literally false by necessary implication “where non-comparableproducts are portrayed as otherwise equivalent (except for the superior orinferior aspect being illustrated in the advertisement).” Interestingly, thecourt actually gave this as part of its instructions, which stated in part:

A statement is literally false bynecessary implication when it does not explicitly state something that isuntrue, but considering the advertisement in its entirety, the only reasonableinterpretation of the statement is that it is untrue. Advertisements using an “applesto oranges” comparison are literally false by necessary implication wherethings that are non comparable are portrayed as otherwise equivalent.

The jury could properly find that Natera’s advertisingstatement was false because it “omits differences which would have beenmaterial to recipients.” And the metrics in the ad, as the ad itself said, were“key” to the products’ functions and therefore to purchase decisions. “Indeed,the fact that Natera highlighted these metrics in its advertising campaignagainst Reveal suggests Natera is being a bit disingenuous in now assertingtheir metrics are not really material.”

Consistent with the amount of money at stake, Guardant alsoprovided a survey to show that the ads affected respondents’ perception of productquality (69.7% of oncologists “understood the main message of Natera’s emailadvertisement to be that Signatera is superior to Guardant’s Reveal”), as wellas evidence that ads of this type affect doctors’ purchasing decisions. Theaccuracy of tests like these ones “undoubtedly” concerned an inherent quality orcharacteristic of the product.

Commercial advertising: Natera argued that it was justdisseminating educational materials, not ads. But the evidence at trial showedthat their purpose was to influence consumers. The comparison chart was“provided to the sales force” and “used by the sales force in meetings withoncologists and physicians for the purpose of influencing customers(oncologists and physicians) to buy its tests.” The ads were widelydisseminated in the relevant market, through thousands of emails. And Guardantpresented evidence that Natera made its false claims with the express economicmotive to advance the sale of Natera’s test over Guardant’s. “The ‘education’here was intended to drive sales.”

First Amendment: A false ad is not protected speech. Nateraargued that, because the comparison advertisement consisted of results from twoscientific studies, ONY, Inc. v. Cornerstone Therapeutics, Inc., 720 F.3d 490 (2dCir. 2013), precluded liability and required treating the ad as a statement ofopinion in a scientific debate. “ONY does not categorically immunizefalse statements about peer-reviewed studies.”  Instead, “[s]tatements in peer-reviewed,published scientific articles are entitled to protection against Lanham Actliability” because, per ONY, they’re opinion rather than fact.  Specifically, for peer-reviewed scientificstudies, “conclusions from non-fraudulent data, based on accurate descriptionsof the data and methodology underlying those conclusions, [and] on subjectsabout which there is legitimate ongoing scientific disagreement ... are notgrounds for a claim of false advertising under the Lanham Act.”

However, “statements made within the academic literature anddirected at the scientific community” are distinguishable from advertising“statements made in commercial advertisements and directed at customers.” [Thedifference here from the Fifth Circuit’s Plastipure decision, on whichthe court (I think correctly) relies, is that the ads in Plastipure werenot directed to doctors. But they were directed to specialist consumers, andneither doctors nor other highly trained professionals are necessarily good atspotting problems in ads that don’t give the full context of a study. (I don’tthink they’re necessarily good at spotting problems in full studies either, butthe Second Circuit’s holding really does need to be confined to that scenarioor, as everyone recognizes, advertising regulation would collapse.)]

Here, there was substantial trialevidence that Natera did not merely reproduce and then accurately describe theresults of a scientific article. Instead, Natera (not scientists in apeer-reviewed, published scientific article) placed, what the jury found to bemisleading, apples-to-oranges comparisons of the studies into a side-by-sideformat.

The instruction on this point included:

Statements in a commercialadvertisement or promotion which are based on test results from apeer-reviewed, published scientific study cannot be literally false unless thatparty challenging the advertisement proves that:

1. The statement in theadvertisement of promotion is not, on its face, supported by the peer-reviewed,published science study. In other words, even if the study is reliable, it doesnot establish the statement at issue in the advertisement; or

2. The statement conveys a falsemessage that is beyond the scope of the peer-reviewed, published scientificstudy, such as comparing test results from a different study when the resultsare not actually comparable; or

3. The statement is supported bythe peer-reviewed, published scientific study but the results of the study werefabricated or fraudulently created.

A party also may show that acommercial advertisement or promotion that relies on a peer reviewed, publishedscientific study is misleading if it reports results from the study in a waythat is deceptive and that deceived a significant portion of the commercialaudience.

The court also found that the jury was properly instructedon awarding corrective damages. As a direct competitor (in fact, the other playerin a two-player market), Guardant was entitled to a presumption of commercialinjury once the jury found material misleadingness. And it also presented “substantialevidence that Natera’s advertisements affected Reveal’s sales and that Guardantneeded prospective corrective damages to remedy the past wrong.” Even though Guardant’stest at the time of the ads was a version no longer on the market, the ads targetedGuardant’s products in general as “tumor-naïve” tests, and the next generationwas also “tumor-naïve.” Given that it was the immediate successor to thetargeted test and “reasonably proximate in time” to the ads, harm wasplausible.

Guardant’s evidence also reasonably supported the jury’sfinding of willfulness. I didn’t go back to see when these statements weremade, but apparently one doctor told Natera that the comparison between the twostudies would be “unfair” and characterized it as “a bit like comparing applesand pears.” In addition, general competitive statements supported the willfulnessfinding with statements such as: “We need to be laser focused … or we will loseto Guardant. We need to put more intensity – this is a war we are entering,” “weneed to go to the mat here” and “[s]pend whatever is necessary to salt[Guardant’s] launch” of Reveal. Testimony indicated that Natera sent about 10emails to each oncologist.

Punitive damages were available on the California claims. The$175 million in punitive damages worked out to a 2.3x multiplier for the damages,“well within the range of an appropriate ratio.”

The court also declined to grant remittitur of the damagesaward; the jury accepted Guardant’s expert’s 3x multiplier for $75 million in prospectivecorrective damages based on the $24.8 million Natera spent on its anti-Revealcampaign.

The court unsurprisingly also granted a permanentinjunction, though it refused Guardant’s request to require Natera to notifycustomers that a jury unanimously found its advertising comparing theperformance of Signatera and Natera was false. “The jury’s award of prospectivecorrective damages is an adequate legal remedy for this request, thus nofurther equitable remedies will be entered.” But Natera was enjoined fromcomparing the tests in terms of the advertised qualities “based, in whole or inpart,” on the two studies at issue.

The jury recommended $42 million in disgorgement, which wasabout 44% of Guardant’s request. The court crunched the numbers on its own, asis appropriate, and found attributing 50% of Natera’s sales of Signatera to thefalse advertising to be reasonable, given that Natera exceeded its projectedsales by about 40-60% and Guardant’s projections were short by about 50%. Thatled to disgorgement of a bit over $37 million. Disgorgement was appropriate toprevent Natera’s unjust enrichment; willfulness supported disgorgement as well.

Guardant also sought a 1.5x multiplier for the actualdamages of $75 million. The Lanham Act allows a court to adjust a Lanham Actaward if it finds “that the amount of the recovery based on profits is eitherinadequate or excessive” up to 3x. But actual damages should “constitutecompensation and not a penalty.” How to square these dueling goals? “Courtssometimes award treble damages because economic harm is hard to prove, wherethere is loss to reputation and goodwill, and to deter future infringingconduct.” Here, though, there was only vague testimony about “the damages thathave been done to the reputation of this product,” and Guardant’s damagesexpert testified that $75 million was “necessary in order to put [Guardant]back in the position they should have been, level the playing field,” and“correct the impressions that have been left in the market.” The juryapparently accepted this testimony. “[A]n assertion by a CEO that additionaldamages should be awarded for e.g. loss of goodwill, without any furtherevidence, does not warrant an additional enhancement of the jury’s award of $75million for corrective advertising.” This was especially true given thepunitive damages award under state law. Willfulness alone didn’t justify amultiplier where “the actual damages awarded already compensate for the allegedharm.”

Prejudgment interest: Section 1117(a) doesn’t mention prejudgmentinterest, but 1117(b) (treble damages for counterfeits) does. And:

In Y.Y.G.M. SA v. Redbubble, Inc.,75 F.4th 995, 1008 (9th Cir. 2023), the Ninth Circuit explicitly noted thatthere is an express provision awarding prejudgment interest in Section 1117(b),but not in Section 1117(c) [statutory damages for counterfeits]. The Courtapplied traditional statutory interpretation tools to find this omission wasintentional by Congress, and thus prejudgment interest was not available underSection 1117(c).

[Ah, if only the Supreme Court had noticed a similar issuewith the noncommercial use exception to dilution’s lack of a “use as a mark”limitation! Anyway.] Expressio unius est exclusio alterius: “When Congressincludes particular language in one section of a statute but omits it from aneighbor, we normally understand that difference in language to convey adifference in meaning.” The same logic applies to regular infringement damages/profits.Also, prejudgment interest is supposed to make plaintiffs whole, but thedamages here were for prospective corrective advertising. And disgorgement alsodoesn’t need prejudgment interest to make a plaintiff whole. “Guardant was notawarded monetary damages for past lost profits, profits that it ‘lost [the]opportunity to invest.’”

The court also denied Guardant’s motion for $22 million inattorneys’ fees. Although the jury found the false advertising willful, thiswasn’t an exceptional case. The key factors are (A) frivolousness; (B)objective unreasonableness; and (C) considerations of compensation anddeterrence. “Natera’s litigation positions were not so unreasonable as torender this case exceptional.” Summary judgment was granted to Natera in someparts of the case, to Guardant on others, and mostly the case went to trial ondisputed issues. “[A] well-fought case where one side wins on all claims doesnot transform a Lanham Act case into one that is exceptional. Were that thecase, attorneys’ fees would be awarded in almost all Lanham Act cases…. Successat a jury trial does not deem the opposing party’s positions weak or frivolous.”

As for litigation conduct, Natera got sanctioned already forcertain conduct, and no further sanctions were warranted. Nor was itexceptional to keep advertising after Guardant sent a cease-and-desist letter; Guardantsued six days later.