Sunday, April 05, 2009

Mammoth opinion in drug false advertising case

Bracco Diagnostics, Inc. v. Amersham Health, Inc., 2009 WL 806581 (D.N.J.)

Fair warning: this decision runs over 120 pages in Word. It is the product of a 39-day bench trial; I hope everyone took a vacation afterwards. Lots of stuff happened, probably the most wasteful of which was that a bunch of expensive expert opinions got excluded on Daubert grounds. The opinion also suggests some of the huge burdens of a false advertising case where many relevant representations are made by sales reps. Here, there were detailed notes of over 300,000 sales visits, creating an enormous job of reviewing and coding them, because the parties disagreed vigorously about how to characterize the notes of what got said.

The parties sued each other for false advertising. Bracco proved some false advertising, but couldn’t establish a causal nexus between the false ads and its alleged lost profits. “[T]he greater number” of Amersham’s ads were true, though some were false in extrapolating beyond the results of reliable, well-controlled studies. Thus, the court granted an injunction and damages for millions of dollars in corrective advertising. On the counterclaims, Amersham gave up on its pursuit of damages and Bracco discontinued the claims at issue, so even though some of Bracco’s ads were false the court didn’t grant an injunction against Bracco. The court required the parties to use an alternative dispute resolution mechanism for future allegations of false advertising.

Falsity

The parties sell x-ray contrast media, which are classified by osmolality. Bracco makes Isovue and Amersham makes Visipaque (a product with the osmolality of blood, also called iso-osmolar). Amersham made establishment claims that its iso-osmolar Visipaque performed better in terms of renal effects than low-osmolar contrast media (LOCM) such as Isovue. Amersham generalized to the class, even though the reliable studies were performed using its own LOCM as the comparator, in part to minimize the impact of its claims on that product, Omnipaque. (There were a bunch of other studies the court concluded did not support the general superiority claim.) In fact, there were no head-to-head studies of Visipaque versus multiple LOCM, or versus LOCM combined with pretreatments, which can improve the performance of LOCM. The FDA has sent warning letters to Amersham indicating that the results of its study can’t be extrapolated to media other than Omnipaque in advertising.

Here’s a bad fact: while one key study that was supposed to prove Visipaque’s renal superiority was ongoing, Amersham took “secret and forbidden peeks at the data looking for trends, and even changed the study endpoints and stopped the study early in response.” When a New England Journal of Medicine reviewer asked if there had been an interim analysis, Amersham and the authors denied it, then amended the article to include this false denial. The “secret attempt to ‘fix’ the study midstream” made the trial non-prospective. The court was also not impressed by the fact that Amersham’s marketing director “provided input to the NEJM article to try to make it misleading, and then celebrated the final version’s obscuring of the limitation of the results of the study to Omnipaque and its overly broad and unsupportable conclusion.”

However, the court determined that the study results were not entirely vitiated by these and other flaws. As long as Amersham carefully identified that Visipaque had been compared to Omnipaque alone, it could use the study results in ads; it just couldn’t use the study to claim general superiority. Notably, the court instructed that if Amersham uses its brand name Visipaque in ads, it must similarly refer to the brand name of the comparator—Omnipaque—instead of the clinical name (iohexol).

The case reminds us strongly that medical literature can say things that the FDA doesn’t allow drug manufacturers to say. One study author testified that he, his co-authors, and the NEJM believed in the scientific reasonableness of the conclusion (extrapolating superiority over Omnipaque to general superiority). But the FDA has repeatedly rejected Amersham’s attempts to make that claim itself.

Bracco doesn’t come off particularly awesome here either; it sponsored a head-to-head randomized trial comparing Visipaque and Isovue on the incidence of cardiac adverse events, then tried to reanalyze the data, undermine the results, and pressure the study’s author when it showed superiority for Visipaque.

Separately, Amersham overclaimed Visipaque’s advantage in patient comfort; its claims are only supported in regard to peripheral angiopathy procedures, not any others, as the FDA has reminded Amersham. Amersham also made unsupported claims of lower costs, based on the idea that Visipaque led to lower rates of adverse events, which are costly. There’s some indication that Visipaque has a higher rate of delayed adverse events than Isovue and other similar contrast media.

Only an unambiguous message can be literally false. The court found three key renal statements: “Visipaque may be better than a LOCM,” “Visipaque is better than all LOCM,” and “Visipaque is as good as or better than a LOCM with prophylactics.” The last two were unambiguous. The first claim was paraphrased from the key study and, without context, may be misleading—in context, it’s clear that “a LOCM” is “Omnipaque,” the LOCM tested. But the court found the phrase misleading on its own. (Is this a stealth application of falsity by necessary implication? I think the doctrine fits.) As for the non-renal superiority claims, there were cardiovascular superiority claims, which included claims of less pain, and cost superiority claims.

These were all establishment claims. (Amersham argued vainly that Bracco had the burden of proof to show by surveys that its claims were establishment claims. It was clear that these superiority claims were based on tests.) So Bracco needed to show that the tests didn’t support the claim, either by attacking the reliability of the tests or by showing that the claim wasn’t sufficiently founded in the tests. The court found that “Visipaque is better than all LOCM” was an extrapolation that strayed too far from the results of the underlying studies. The court found that those studies were reliable, but without identifying which LOCM was tested, disseminating the studies’ conclusions was misleading. Amersham must clearly and conspicuously state which drugs were actually tested. And it may not explicitly or implicitly communicate a superiority message over Isovue without a reliable head-to-head study. The non-renal superiority claims fared similarly.

The court found testimony about FDA guidelines to be legally irrelevant. In particular, the FDA requires substantial evidence before allowing superiority claims, defined as two or more adequate and well-controlled studies directly comparing the same products. But it’s not sufficient under the Lanham Act to show that claims are inadequately substantiated under FDA rules; claims must be shown to be literally false or misleading. Thus, a defendant could advertise the results of one adequate well-controlled head-to-head study without violating the Lanham Act (and just wait for the FDA letter, I suppose).

However, the FDA’s response to the claims at issue is probative and persuasive. Here, the court didn’t have to interpret the FDCA or speculate on the FDA’s position, since the FDA had provided its views in “numerous letters” to Amersham about Visipaque’s misleading superiority claims. (One lesson: even if FDA enforcement is dead, worry about your competitors’ use of FDA warning letters!) Amersham argued that, to the contrary, the FDA never acted against Amersham, despite Bracco’s attempts to convince it to do so, and that Bracco was using the Lanham Act as an end run around the FDA’s inaction, since none of the FDA letters were final agency action.

But the court wasn’t usurping the FDA’s authority or preempting its findings. The FDA had already made its position abundantly clear. Still, Bracco can’t prove falsity simply by relying on the FDA’s letters applying FDA’s standards. Inadequate substantiation under FDA standards is not the same as falsity. But Bracco had other evidence of falsity.

On implied falsity, the court excluded Bracco’s consumer survey and thus had no evidence of misleadingness, as required by the case law.

Amersham’s falsity was not willful, because the claims were based on scientific studies that have not been invalidated—just not sufficiently based. Amersham did have a protocol in place attempting to weed out false claims in ads, but, perhaps unsurprisingly, that protocol let through some claims that a neutral adjudicator later found false.

Puffery

I’m not sure why the court thought it necessary to say this, but: Amersham’s statements were specific and measurable factual claims, not puffery. The use of footnotes also didn’t help, because footnotes purporting to change the apparent meaning of claims but are inconspicuous or in fine print can’t remedy a misleading claim. Amersham’s footnotes simply cited studies, without further explanation and in particularly without explaining the studies’ limitations.

Materiality

The court found materiality because: (1) Amersham’s sales and marketing teams repeated certain unsupported claims; (2) the substance of the claims can’t be observed anecdotally by doctors (this is not directly about materiality, but it shows that doctors would have to take the claims on faith); (3) the type of claims—drug safety—were shown by Amersham’s own observations and market research to be useful at generating sales; (4) there was evidence that it will take several years for Bracco to recover from the falsity. Also, given that some of the claims were literally false, there is a presumption of materiality and deception.

Commercial Speech

Not everything at issue in the case was commercial speech actionable under the Lanham Act. Most obviously, internal company documents that were never publicly disseminated in the US weren’t actionable.

Scientific articles in peer reviewed journals also aren’t commercial speech, even if the underlying research was funded by one of the parties in the hopes of improving its economic position. Amersham may have sponsored the research whose results were reported in the NEJM, but it wasn’t the author; the author wasn’t paid for his work on the article; the article didn’t advocate a particular purchase, even though it came to specific conclusions about which product was better suited for medical purposes; and the article was published by the NEJM, which is an impartial educational journal.

In its initial form, the court refused to inquire into the article’s reliability, but the court nonetheless determined that using articles “in a secondary dissemination in the form of commercial advertising” puts their reliability in question. Manufacturers will only seek to disseminate favorable information, and their considerable financial resources mean that favorable findings are more likely to reach doctors than unfavorable findings, distorting doctors’ understanding of safety and efficacy. When manufacturers disseminate favorable studies in order to get people to buy their drugs, they are engaging in commercial speech.

Likewise, oral statements made by Amersham’s sales reps were actionable commercial advertising or promotion, if proved. Amersham argued that the sales notes weren’t sufficient evidence of what the sales reps said; and as mentioned above, there were epic disputes about interpreting whether the notes indicated any kind of falsity. But many cases accept sales notes as evidence of what was said. Under the standard Lanham Act test, commercial advertising or promotion must be disseminated sufficiently to the relevant purchasing public to constitute advertising or promotion within that industry, which in the case of oral statements means that they’re widely disseminated and part of an organized campaign in the relevant market.

Amersham argued that the number of actually false statements at issue was too small to be actionable. Bracco contended that, given how vital face-to-face communication is to selling drugs, the sales call notes were enough. The court agreed: the sales calls weren’t made in isolation, but as part of a large-scale marketing plan. “[E]ven if the offending sales calls are a very small percentage, and thus, alone would not be actionable, when the sales calls are combined with [the] overall campaign, which was promoted through press releases, websites, and [continuing medical education], the result is false ads which have been sufficiently disseminated to be actionable under the Lanham Act.”

Amersham also argued that a good faith effort by a company to educate its sales force about what can fairly be said about published studies rendered any limited false or misleading statements made by representatives outside of those parameters not actionable. But that’s not the law; the law targets not unreasonableness, but falsity. Nonetheless, the court considered Amersham’s training “evidential,” just not dispositive; I’m not clear what that means, but my guess is it has to do with the ultimate finding that Bracco couldn’t show that it was the falsity that cost it sales.

As for the rest of it: website ads, print ads, and TV ads are all actionable, as were continuing medical education events (CMEs) sponsored by Amersham, because they were designed by Amersham to deliver a specific message of Visipaque’s renal superiority. CMEs are presented by doctors, but here Amersham had a substantial role in the creation of their content. One example of a sponsored CME used slides that had the defendant’s logo on every slide and was given by an authorized representative. The “direct control” made the CME commercial. The court also specifically found commercial speech when defendant’s logo was on the first slide of another presentation and the presentation was made in part by its paid consultants.

Unclean Hands

Bracco’s own alleged false advertising wasn’t a barrier to relief because Amersham’s claims had significant safety implications. The unclean hands defense is rarely successful, because of a strong public interest in avoiding misleading ads. Bracco’s misconduct, even if taken as established, was largely defensive and limited in scope and duration.

Injunctive Relief

Bracco met the standards for injunctive relief. The court ordered corrective advertising: Amersham is to issue a press release, including on its website, about the decision and the corrective ads. Corrective advertising is appropriate in cases of falsity that bear on public health. Amersham is also to re-train sales and marketing personnel in accordance with the opinion. Future disputes about Amersham’s ads are to go to arbitration, which may include NAD (though I suspect NAD might decline jurisdiction over some standard methods of communication with doctors, like CME and sales rep visits), and Amersham is to bear the costs associated with the disputes if the ads are found to be false, while Bracco will bear them if the ads are not found to be false.

Damages

Falsity leads to a presumption of consumer deception and harm. But that doesn’t alone justify an award of money damages. For that, you need willfulness or other evidence. Here, there was insufficient evidence of willfulness; the studies were subject to interpretation (though Amersham wrongly interpreted them in its own favor) and Amersham did have an extensive ad approval process that considered FDA and other issues.

Amersham’s witnesses conceded that the Visipaque renal safety claims drove substantial sales increases, a big chunk of which were at Bracco’s expense, since the market is effectively a duopoly. Nonetheless, only some of the claims were false; others were “accurate touting of favorable results of reliable scientific studies,” which means it was impossible to conclude that the false ads were the cause of Bracco’s lost profits, so lost profits couldn’t be awarded. In particular, the court found that the falsity wasn’t the cause of Bracco losing several major contracts to Amersham. The basic problem was that some of Amersham’s uniqueness claims were true, while others were false, and it’s hard to tell which drove sales.

Disgorgement of Amersham’s profits was also inappropriate. Using the Third Circuit’s five-factor test: (1) Amersham’s acts weren’t willful or deliberate. (2) The court found insufficient evidence of sales diversion, as noted above. (3) Other remedies—injunctive relief and compensatory damages for the costs of Bracco’s corrective ads—were sufficient. (4) Bracco acted fast in asserting its rights, which favors disgorgement. (5) There’s a strong public policy in making false advertising unprofitable, deterring false statements about drug safety, and deterring inflated drug prices based on inflated clinical claims. But deterrence alone isn’t enough to justify disgorgement without willful conduct.

Bracco was, however, entitled to recover its costs for corrective ads and other damage control expenses if (1) there was a likelihood of confusion or damage to sales, profits or goodwill; (2) the damage control expenses were responsive to the misconduct; and (3) the expenses were reasonable under the circumstances and proportionate to the damage that was likely to occur. Bracco met these requirements. Hundreds of millions of dollars were at stake; it was reasonable of Bracco to spend a bunch, and the court awarded nearly $11.4 million.

However, the court concluded that Amersham wouldn’t have to pay Bracco’s expenses in investing in research to disprove the falsity. The parties compete fiercely, and Bracco has an incentive to invest in studies anyway. Given the number of studies submitted to the court, it appears to be “commonplace, if not a necessary part of the industry,” to spend significant amounts on comparative studies. Indeed, virtually every clinical trial, study, and publication in the area was sponsored by one of the parties; the court lamented the absence of a truly independent clinical study.

Bracco also didn’t get attorneys’ fees, which probably ate up a huge amount of the award (one expert alone got paid $500,000, which among other things made the court skeptical of claims that Bracco couldn’t have done a bigger survey because of the expense).

Counterclaims

Amersham alleged that Bracco falsely advertised Isovue’s superiority to Omnipaque. The court agreed that Bracco’s tests didn’t prove its claims. Bracco stipulated that these claims were no longer in use, and the court found no real possibility that it would revive the claims, so it declined to grant injunctive relief. But if disputes do arise, the same arbitration procedure is to be used, with the same cost-bearing structure.

No comments: