The Duty to Warn in the Age of Pharmacogenomics

A $700 million settlement in Hawaii and ongoing litigation in Texas have brought pharmacogenomic evidence, genetic susceptibility, and the pharmaceutical duty to warn to the forefront of product liability litigation.

Part 1: Plavix Product Liability Litigation in Hawaii and Texas

Introduction: Genetic Evidence and the Pharmaceutical Duty to Warn

Historically, pharmaceutical product liability litigation has focused on whether drug manufacturers adequately warned patients and prescribing physicians about a medication’s risks. Emerging litigation asks a different question: whether a manufacturer has a duty to warn that a drug may provide less benefit to a genetically identifiable subgroup of patients.

Two state attorney general actions against the makers of Plavix (clopidogrel), one ultimately resulting in a $700 million settlement in Hawaii and the other ongoing in Texas, place pharmacogenomics and genetic susceptibility at the center of that question. The allegations focus on what the manufacturers knew, or should have known, about genetic variation affecting clopidogrel metabolism and clinical response.

As discussed below, no single rule determines when developing scientific evidence becomes sufficient to require a drug label update. The inquiry is fact specific and depends on the governing law, the evidence available at the relevant time, and what the manufacturer knew or reasonably should have known. Attorneys handling pharmaceutical product liability matters involving genetic evidence therefore need a working understanding of pharmacogenomics and genetic susceptibility.

This three part article will (1) examine the Hawaii and Texas Plavix litigation, (2) provide a brief genetics and pharmacogenomics primer, and (3) explain how genetic susceptibility could be used in product liability litigation, taking Plavix as an example.

Plavix, Clopidogrel Metabolism, and CYP2C19 Genetic Variants

Plavix (active ingredient: clopidogrel) is an antiplatelet (i.e., “clot preventing”) medication prescribed to patients with cardiovascular disease to reduce the risk of heart attacks and strokes by preventing blood clots (Lee et al., 2022). It became one of the world’s best selling prescription drugs; the State’s complaint reported U.S. sales of $6.62 billion in 2011.¹ The 2024 retrial order found that approximately 834,012 Plavix prescriptions, refills, and non-retail units were sold in Hawaii between December 1998 and March 12, 2010.²

Clopidogrel is a prodrug: it must be metabolized into an active form before it can produce its intended antiplatelet effect (Kazui et al., 2010). CYP2C19 is one of several enzymes involved in this bioactivation pathway (Patel et al., 2025).³

At this point, many lawyers may pause: what is CYP2C19, and how is it pronounced? It is commonly pronounced “sip two see nineteen.” Part 3 explains the nomenclature and biology in greater detail. For now, the key point is that CYP2C19 refers to a gene and the enzyme it encodes, which plays an important role in converting clopidogrel into its active form.

There are multiple CYP2C19 genetic variants, some of which reduce or eliminate enzyme function. Patients with an intermediate or poor metabolizer phenotype may form less active clopidogrel metabolite and experience reduced platelet inhibition. In cardiovascular settings where genotype-guided therapy has been studied, that reduced response can be associated with a higher risk of adverse cardiovascular or cerebrovascular events (Lee et al., 2022). This relationship between genetic variation and drug response is a central pharmacogenomic issue in the Plavix litigation and is discussed in greater detail in Part 3.

Researchers identified CYP2C19 genetic polymorphisms in the 1990s through studies of mephenytoin metabolism (De Morais et al., 1994). Their relevance to clopidogrel emerged later. Early studies beginning in 2006 linked loss-of-function alleles with reduced platelet response to clopidogrel (Hulot et al., 2006). In 2009, larger clinical studies reported associations among CYP2C19 variants, reduced clopidogrel response, and major adverse cardiovascular events (Mega et al., 2009; Simon et al., 2009). On March 12, 2010, the FDA added a boxed warning explaining that CYP2C19 poor metabolizers form less active metabolites and have diminished antiplatelet response; the agency advised healthcare professionals to consider alternative treatment or dosing strategies (U.S. Food and Drug Administration, 2010).

The Hawaii Plavix Litigation

The State of Hawaiʻi’s Allegations

The State of Hawaiʻi, through its Attorney General, brought an Unfair or Deceptive Acts or Practices (UDAP) enforcement action against the manufacturers of Plavix in state court in 2014.⁴ Hawaii’s central theory was that the manufacturers failed to disclose material information about reduced efficacy and related risks for a genetically identifiable subset of patients, despite allegedly knowing or having reason to know of the issue.

More specifically, the State of Hawaii alleged:⁵

• The manufacturers allegedly ignored, concealed, or minimized clinical trial data and other information suggesting that Plavix was no more effective than aspirin for certain patients and could present gastrointestinal bleeding and other risks.
• They allegedly marketed Plavix as safer and more effective than aspirin and competing drugs despite that information.
• They allegedly promoted Plavix for uses not shown to be safe or effective and caused false reimbursement claims to be submitted for prescriptions that were not medically necessary or cost effective under Hawaii law.

The Manufacturers’ Defense: Scientific Knowledge and Drug Labeling

The central question in the Hawaii litigation was when pharmacogenomic evidence became sufficiently developed to trigger a manufacturer’s duty to warn or update the Plavix label. The Hawaii decisions do not establish a bright-line rule. Instead, the courts examined the timing of published studies, internal company documents, and evidence concerning what the manufacturers knew or should have known. Because pharmacogenomic evidence often develops incrementally, from early associations to replicated findings and clinical guidance, the analysis is highly fact-specific and requires both legal analysis and scientific literacy.

The manufacturers raised two arguments against liability, among other things:

1. Timing and Knowledge: Defendants argued that liability could not be based on scientific information unavailable to them or the broader scientific community at the relevant time. They contended the record did not establish sufficient knowledge of clinically meaningful variability in clopidogrel response to support the earlier label language Hawaii claimed was required.
2. Materiality: Separately, Defendants argued that even if genetic susceptibility to variable clopidogrel metabolism exists, CYP2C19’s contribution is not significant enough to be material. This rested on several points: that CYP2C19 loss-of-function alleles account for only about 12% of overall response variability; that multiple other enzymes are also involved in clopidogrel activation; and that certain studies (CHARISMA, COMMIT) did not show a meaningful CYP2C19 outcome link.

Verdicts, Appeals, and the $700 Million Hawaii Settlement

The case was tried before Judge Dean E. Ochiai without a jury over four weeks in October and November 2020. On February 15, 2021, the trial court found that the manufacturers had engaged in unfair and deceptive practices in Hawaii from Plavix’s launch through the addition of the boxed warning in 2010.⁶ Its findings included a failure to update the warning after learning of reduced response in poor metabolizers and a failure to pursue research that could have “negative marketing implications.”⁷

On appeal, the Hawaiʻi Supreme Court affirmed the judgment on the State’s unfairness claim but held that the trial court had erred by resolving materiality on summary judgment for the deceptive practices claim. It vacated the deceptive practices judgment and civil penalties and remanded those issues for a new trial. The remand trial took place before First Circuit Court Judge James H. Ashford from September 25 through October 16, 2023. On May 21, 2024, the court entered findings and a $916,012,000 judgment.⁸

The manufacturers appealed again. Before that appeal was resolved, the parties entered a $700 million settlement on May 9, 2025, with Bristol-Myers Squibb and Sanofi each agreeing to pay $350 million. The settlement agreements state that the manufacturers continued to dispute the claims and denied liability. The State’s release described the agreement as ending nearly 12 years of litigation, with payment due by June 9, 2025.

The Texas Plavix Litigation

On November 20, 2025, Texas Attorney General Ken Paxton filed suit against Bristol-Myers Squibb and Sanofi in the 71st District Court of Harrison County.⁹ The Texas complaint alleges violations of the Texas Deceptive Trade Practices Consumer Protection Act (DTPA) and the Texas Health Care Program Fraud Prevention Act (THFPA). According to the complaint, the companies failed to warn patients and prescribing physicians that Plavix could have diminished or no therapeutic effect for many patients, with particular emphasis on Black, East Asian, and Pacific Islander patients because of alleged differences in the prevalence of genetic variants affecting CYP2C19 function.¹⁰

The Texas case advances a related theory under Texas statutes, testing the alleged relationship among genetic susceptibility, clopidogrel effectiveness, and manufacturer warnings outside Hawaii’s UDAP framework. Those allegations have not yet been resolved on the merits.

The defendants removed the case to the U.S. District Court for the Eastern District of Texas (Case No. 2:25-cv-01180). On March 12, 2026, the federal court adopted the magistrate judge’s report and recommendation and remanded the matter to the 71st District Court of Harrison County.¹¹

The timing and knowledge and materiality disputes are central to the theories advanced in both matters. The relevant questions include, for example, whether a manufacturer had a duty to warn may depend on when the underlying pharmacogenomic evidence became sufficiently clear to require disclosure, including how CYP2C19 affects clopidogrel metabolism, how the relevant genetic variability was identified, and how confidently that variability could be linked to clinical outcomes. Separately, materiality asks whether that information, once known, was significant enough to affect a reasonable consumer’s or physician’s treatment decisions. That question depends, for example, on the clinical weight of the variability itself, including how large its contribution to patient risk was relative to other factors affecting drug response. Because this science developed incrementally, and because the field involves inherent complexity, different fact finders may assess its significance differently. Understanding the underlying science is therefore essential both for the lawyers presenting the evidence and for the judge or jury evaluating what individual studies showed at the relevant time. Parts 2 and 3 provide the scientific background needed to assess those questions.