Eli Lilly v. Novartis (Kisqali)

Melissa C. Brown*

I. Why It Made the List

On June 3, 2022, Novartis released the results of its latest clinical trial for its breast cancer treatment, Kisqali. The headline read:

New CDK4/6i data at ASCO reinforce Novartis Kisqali as only drug in class with consistently proven overall survival benefit in HR+/HER2- metastatic breast cancer.

The announcement had special resonance in the industry as the following day, Pfizer announced that the results of the most recent trial of its competing drug Ibrance, although promising, did not reach statistical significance. On the back of the announcement, Novartis launched a broad national advertising campaign directed to both health care providers and consumers, including a television commercial which aired on several major networks, website advertising, and physician brochures, fliers and conference presentations touting the results of it studies. The following month, after a brief and unsuccessful round of correspondence with Novartis, competitor Eli Lilly filed a challenge with the National Advertising Division of BBB National Programs (NAD), alleging that Novartis’ advertising made false and misleading claims about its Kisqali drug.

The decision that issued at the end of that year is notable as a casebook study of the challenges of substantiating health benefit claims and the limitations of cross-trial comparisons submitted for that purpose. It is also a striking illustration of how an advertising claim directed to health care professionals can convey a distinctly different message to the lay public and goes to the heart of the issues that arise in advertising prescription drugs—especially sophisticated treatment drugs—directly to consumers.

II. Discussion

Metastatic breast cancer, or stage IV breast cancer, is cancer that has spread beyond the breast to other parts of the body. Roughly 20–30% of persons diagnosed with early breast cancer will develop metastatic breast cancer, the most common form of which is HR+, HER2 breast cancer. Metastatic breast cancer is presently incurable, but current treatments can reduce the spread of cancer to other parts of the body, consequently extending time without disease progression and enabling patients to live longer—an outcome referred to as “overall survival.” Kisqali belongs to a class of treatments known as CDK 4/6 inhibitors. CDK 4/6 inhibitors block proteins of the same name found in both healthy and cancer cells and which, in breast cancer, can become overactive and cause the cells to grow and divide uncontrollably and precipitate tumor growth. In the United States, there are currently three CDK4/6 inhibitors that have been approved by the FDA to treat HR+, HER2 breast cancer: Pfizer’s Ibrance, Lilly’s Verzenio and Novartis’ Kisqali. Each drug has been the subject of several significant phase III clinical trials with overall survival as a key clinical endpoint. The test groups in each study received the target drug in combination with an adjuvant therapy (usually an endocrine therapy), and survival outcomes were compared against the placebo group taking the same therapy alone.

Both parties submitted data as to their clinical trials.[1] In addition, Novartis introduced a meta-analysis of global CDK4/6 research to date. That analysis indicated that no other CDK4/6 treatment had achieved statistical significance in three phase III trials. The claims at issue in Novartis were grounded in this clinical landscape.

Specifically, Lilly challenged the headline claim in the press release and several variations of the express claim “KISQALI – the longest median overall survival ever reported in HR+/HER2 mBC,”[2] along with the implied claims that Kisqali provided superior survival benefits over competing treatments and that patients taking Kisqali would live longer than patients taking any other comparable drug. During the course of the proceedings, Novartis informed NAD that it was permanently discontinuing all but one of the challenged express claims—a result not uncommon in response to an NAD challenge—as well as the implied claims alleged to flow therefrom. Novartis also withdrew the commercial in which those claims appeared. Remaining for NAD’s review was the claim “The only drug in class with consistently proven overall survival benefit in HR+/HER2- metastatic breast cancer,” (the “only drug in class” claim).

Lilly argued that the remaining claim was undeniably comparative and went beyond merely reporting the results of Novartis’ MONALEESA trials and conveyed the message that Kisqali provides superior survival benefits as to other available treatments, including Verzenio—a message that was unsupported in the absence of head-to-head testing of Kisqali against the market. Lilly also emphasized that Novartis’ advertising targeted a vulnerable population grappling with an incurable disease. Novartis argued that its claim was not a superiority claim and at best a parity claim that was fully supported by current clinical data.

In an NAD proceeding, the advertiser bears the burden of proof and is responsible for all messages reasonably conveyed by the advertising, not merely those it intended to convey. Moreover, although NAD false advertising analysis often evaluates advertising from the perspective of the reasonable consumer, NAD has recognized that the takeaways by discerning professionals may be critically distinct from the interpretations of that same advertising by the lay consumer. NAD therefore evaluated the challenged claim through the lens of each of the target audiences.

i. Consumer-Directed Advertising

As a threshold matter, NAD determined that the claim, “only drug in class with consistently proven overall survival benefit in HR+/HER2 metastatic breast cancer across three phase III trials,” is inherently comparative. By virtue of the phrase “only drug in class”, the claim pits Kisqali against all other CDK4/6 inhibitors and posits it has an attribute that the others do not have—here, a consistently proven survival benefit across three clinical trials.[3] NAD noted that to a lay consumer, a reasonable conclusion from the statement that a drug has been consistently—and more often—proven to achieve a statistically significant survival benefit is that that product is more effective.[4] NAD noted further that as used here, the term “consistently” implies that the results of trials of competing drugs have been inconsistent—a message that, without qualification, was open to negative inferences beyond the mere absence of regularly achieving a statistically significant result. As such, NAD found that, to the lay audience, Novartis’ quantitative claim could convey a qualitative message. NAD also agreed with Lilly that the target audience here would be especially susceptible to claims of improved survival.

NAD reasoned that most consumers do not have the medical knowledge or experience to understand the nuances of clinical trials. They may not appreciate that the results of a clinical trial may be influenced by a variety of factors apart from the quality and efficacy of the tested drug, such as trial design, patient population characteristics, interactions and nature of adjuvant drugs, etc., and as such, they will not understand that outcomes across trials are difficult to compare. NAD reasoned further that the lay consumer will not understand that achieving statistical significance across three clinical trials does not in and of itself establish superior efficacy to other drugs that have proven effective in a smaller number of trials.

Novartis argued that because Kisqali is a prescription drug, it cannot be obtained directly by the patient, and the oncologist could correct any possible misleading messages conveyed by the challenged claim. However, NAD determined that the fact that a consumer cannot purchase Kisqali by herself or that an oncologist may correct any misinterpretation of Novartis’ advertising did not remove the initial impression of the claim. Nor did it absolve Novartis of its obligations to ensure that the initial impression of its advertising was not misleading.

In sum, NAD concluded that one message reasonably conveyed to consumers by the “only drug in class” claim is that Kisqali is more effective and provides superior survival benefits to other drugs in its class, including Verzenio, and that patients taking Kisqali will live longer than when taking any other CDK4/6 treatment. NAD also concluded that Novartis’ evidence did not support such a claim. NAD noted first that clinically proven health claims are held to a very a high standard of proof and that where express or implied superiority claims are at issue, head-to-head testing is the most reliable form of substantiation. NAD also noted that data accumulated from different tests cannot be reliably compared unless it is established that the data resulted from tests that were “essentially identical or all of the variables are accounted for.”[5]

NAD found that the clinical data presented by Novartis did not allow for meaningful assessment or comparison. Although that evidence established that Kisqali had achieved the longest median overall survival outcome in a published clinical trial of CDK4/6 inhibitors, Novartis presented no evidence that the methodologies in the studies it had submitted were similar enough to allow NAD to properly compare the reported overall survival data, nor any statistical analysis of those results. Additionally, Novartis did not provide any details as to the patient populations enrolled in each trial or other critical elements of the trials being compared (such as study design, period of follow-up, etc.). Further, as Lilly pointed out, the comparator arms (endocrine therapy combinations) were not the same between the studies relied on by Novartis and submitted to the record. There were also important differences even as between the Lilly and Novartis studies evaluating Verzenio and Kisqali in combination with fulvestrant, especially as to the patient populations enrolled in each trial. For example, MONALEESA-3 enrolled a higher proportion of patients with advanced or metastatic breast cancer who had never received endocrine therapy as treatment for breast cancer (endocrine naïve), while MONARCH-2 excluded these patients from the intent-to-treat population. Additionally, MONARCH-2 enrolled more patients with clinical features suggestive of endocrine resistant disease.

For all these reasons, NAD concluded that the implied superiority messages in the “only drug in class claim” were not supported and recommended that Novartis discontinue the claim in patient-facing advertising and refrain from making any claims that implied that Kisqali provides superior survival benefits to other comparable drugs or that patients taking Kisqali would live longer than patients taking any other available treatment.

ii. Physician-Directed Advertising

NAD reached a different conclusion with respect to Novartis’ physician-directed advertising claim: “The only CDK4/6 inhibitor with statistically significant overall survival proven across 3 phase III trials.” The claim appears in a context surrounded by data, including the parameters of all three of the MONALEESA studies, follow up times, confidence bounds, patient subgroups tested, extensive drug and safety information, and additional notes on toxicities, adverse reactions, lab abnormalities, footnote references, and a graph of MONALEESA-2 data as the study progressed. NAD determined that the target audience—primarily oncologists—would discern that the comparative claim related to clinical data points and did not convey a broader message of relative efficacy. First, NAD reasoned that the oncologist audience would be well versed with the science in this area and familiar with the intricacies of breast cancer research and as such, would be fully equipped to appreciate both the significance and the limitations of the reported data, especially when provided with sufficient detail as to the trials’ design and findings. For the same reasons, NAD determined that the professional audience would understand that, although the achievement of a statistically significant survival outcome in more trials than its competitors is a promising result, it is not conclusive that Kisqali provides superior survival benefits to other comparable drugs or superior survival benefits to all classes of patients with HR+/HER2 metastatic breast cancer.

NAD concluded that clinical experience and the context provided in the HCP advertising would both inform the physician takeaway and limit it to the recited facts, and that this audience would interpret the comparative claim here simply as reporting that Kisqali is unique in achieving a statistically significant overall survival benefit across Novartis’ three phase III clinical trials. NAD also found that this message was amply supported by the advertiser’s evidence.

Novartis agreed to comply with NAD’s recommendations. On the patient section of its redesigned website for Kisqali, the claim now reads: “Proven to Extend Lives in 3 Large Clinical Trials.”

III. Impact

First and foremost, Novartis was a resounding win for consumers, and misleading claims targeting a vulnerable population were removed from the marketplace in a matter of months. Indeed, to patients with terminal cancer, Novartis’ survival claim was the ultimate health benefit claim. Second, Novartis was an affirmation for self-regulation, and a high-stakes dispute was resolved quickly, efficiently, and with limited expenditure of resources. Third, the decision offers important guidance for advertisers seeking to make comparative clinical effect claims or comparative health benefit claims generally. More broadly, NAD has no discretion to reject challenges brought before it and is regularly tasked with addressing novel issues. Thus, although NAD review relies heavily on FTC rulings and guidance—and where regulated products are at issue, seeks to harmonize its decisions with federal regulation—it is often called upon to fill in the gaps of agency rules and guidance.

Finally, and perhaps most critically, Novartis is an example of how NAD can look at claims—especially implied claims that may not rise to the level of priorities of the FDA. Further, for example, Lilly noted in its challenge that in fiscal year 2021, the FDA’s Office of Prescription Drug Promotion received over 145,000 unique submissions through the Form 2253 process. As the decision illustrates, NAD can take a deep dive into these claims and the full spectrum of contexts in which they appear. Clearly, the risks of overstatement in comparative health benefit claims are not limited to prescription medications. However in this arena, with claims about a penultimate health benefit, the  stakes can be much, much higher.

About the National Advertising Division

Founded in 1971 by the U.S advertising industry, the National Advertising Division is the industry’s self-regulatory forum for review of national advertising. NAD reviews national advertising claims directed to consumers, professionals, or business entities, in any media. The majority of cases heard by NAD are advertising challenges brought by competitors. However, through its Monitoring Program, NAD can initiate a challenge based on its own monitoring of the marketplace and review advertising claims in a variety of contexts and product categories. NAD currently issues over 100 decisions each year, and NAD’s appellate arm, the National Advertising Review Board (NARB), currently hears over a dozen cases annually.

In 2023, approximately 25% of NAD cases involved products regulated in whole or in part by the FDA, and every quarter, NAD submits a report of its decisions of cases involving FDA regulated products to various division of that agency. As a coda to the Novartis case, in January of this year, the FDA issued a warning letter to Novartis for a new television commercial making a compound claim that Kisqali helped patients with metastatic breast cancer both live longer and live well.

* Melissa Brown is an attorney with the National Advertising Division of BBB National Programs and handles false advertising cases across a wide range of industries and media. Prior to joining NAD, she was a Visiting Scholar in Intellectual Property at Columbia Law School and then of counsel to the IP and commercial litigation boutique Leichtman Law.

[1]   Lilly’s first published trial, MONARCH-2, studied Verzenio in combination with fulvestrant and yielded a statistically significant improvement in overall survival as compared to the placebo. Its latest study, MONARCH-3 combining Verzenio with an NSAI, is ongoing, and results have not yet reached statistical significance. Novartis has published three clinical trials, MONALEESA-7, in combination with goserelin and either an NSAI or tamoxifen; MONALEESA-3, Kisqali with fulvestrant; and MONALEESA-2, Kisqali with letrozole. In all three trials, Kisqali demonstrated a statistically significant survival benefit over the placebo. Separately, public records indicate that Pfizer had published two clinical trials, PALOMA-3, in combination with fulvestrant, and PALOMA-2, in combination with letrozole. In both trials, although patients in the Ibrance group demonstrated a longer overall survival than patients in the placebo, neither trial had achieved results that were statistically significant.

[2]   The other challenged claims were:

“Live Longer with KISQALI – The Longest Overall Survival Data Ever Reported in HR+, HER2- mBC”;

“The longest survival data ever reported in HR+, HER2- mBC”;

“Kisqali has the longest median overall survival ever reported in HR+/HER2- metastatic breast cancer.”

[3]   Novartis informed NAD that the claim would replace the discontinued “longest survival” claims on the patient-facing section of the Kisqali website.

[4]   See i-Health (Culturelle), Report #6196, NAD/CARU Case Reports (January 2018), involving a challenge to a series of clinically proven claims for i-Health’s Culturelle probiotic supplement, including the claims: “LGG is the most clinically proven effective strain*” “Based on the studies of a range of benefits throughout the lifespan”; “LGG is the most proven effective strain*” “*Based on the number of Lactobacillus rhamnosus GG clinical studies, as of May 2017.” NAD determined that one message reasonably conveyed by the advertiser’s “most proven effective” claims was that the strain of probiotic in Culturelle had been proven to be more effective than competing strains in providing the benefits associated with probiotics.

[5]   See, e.g., Colgate-Palmolive Co. (Colgate Optic White Toothpaste) Report #5490, NAD/CARU Case Reports (July 2012) (disallowing comparison of results of clinical testing on advertiser’s toothpaste to results of testing on challenger’s product where “studies were conducted at different sites, using different protocols, and using different criteria for participation, with different baselines); Unilever US, Inc. (Vaseline Sheer Infusion) Report #5262, NAD/CARU Case Reports (Dec. 2010) (rejecting combination of advertiser’s monadic sensory testing and statistical PCA (Principal Component Analysis) data to support claim advertiser’s product had a “silkier feel” over other lotions); Novus International, Inc. (Mintrex and MAAC Organic Copper Supplements for Livestock), Report #5597, NAD/CARU Case Reports (May 2013) (rejecting series of studies on bovine liver copper values to support superiority claim in the absence of statistical analysis of performance of copper supplements as compared to each other and not to placebo); Procter & Gamble (Crest White Strips) Report #3918, NAD Case Reports (June 2002) (rejecting cross study comparison because of the “many differences between the [two] studies” submitted); Den-Mat Corp. (Rembrandt Plus Superior Bleaching system and Dazzling White Tooth Bleaching Value Kit) Report #3814, NAD Case Reports (Sept. 2001) (disallowing comparison of data from advertiser’s product to multiple tests of challenger’s Crest Whitestrips where “the methodology varied with respect to the number of people being evaluated, the accompanying dentifrice that was used by test participants and inclusion or omission of pre-test prophylaxis . . . made a comparison of the resulting data untenable”); Ecofibers, Inc. (d/b/a Precision Fibers) (Hydroseeding Mulch) Report #3905, NAD/CARU Case Reports (May 2002) (disallowing comparison of results of two separate studies due to differences in methodologies).