This article was published in the Appellate, Consumer Protection, Health, Life Sciences and Product Liability sections of Law360 on March 12, 2015. © Copyright 2015, Portfolio Media, Inc., publisher of Law360. It is republished here with permission.
On Feb. 25, 2015, nearly two years after argument, the South Carolina Supreme Court directed entry of a $136 million judgment against Janssen Pharmaceuticals Inc.1 These civil penalties stemmed from a suit brought by the state's attorney general relating to Risperdal, Janssen’s “atypical,” or “second generation,” anti-psychotic medication, which has been on the market since 1994. The court found Janssen’s communications to doctors had a “tendency to deceive,” even though there was no evidence that any doctor had been deceived or any patient harmed.
The state made two allegations of misconduct by Janssen: (1) its failure to use the Changes Being Effected provision to strengthen the Risperdal label regarding weight gain and metabolic effects (the labeling claim)2 and (2) the sending of a November 2003 Dear Doctor letter that the U.S. Food and Drug Administration's Division of Drug Marketing, Advertising and Communications3 found misleading because it undercut a classwide warning on metabolic effects required by the FDA (the DDL claim). On both claims, the jury found Janssen violated the South Carolina Unfair Trade Practices Act. The trial judge imposed penalties of $327 million, which the court reduced to $136 million.4
Professing to protect patients — none of whom suffered injury — and doctors — none of whom were misled — the South Carolina Supreme Court approved a huge penalty, but gave short shrift to important legal issues, such as federal preemption and the First Amendment. The court’s decision essentially permits the state to act as an enforcer for the FDA through an expansion of state consumer protection laws.
Risperdal, as the court grudgingly acknowledged, is an “effective drug” used to treat serious mental illnesses. The state’s own expert, testifying in another case, stated that “second-generation anti-psychotics are among the most powerful disease modifiers in all of modern medicine and that psychiatrists felt [Risperdal] was a ‘miracle drug’ because it did not have the serious side effects of first-generation anti-psychotics.”5
No Harm, But Huge Penalties
According to the South Carolina Supreme Court, there was “little evidence of actual harm” and the medical community “was aware of the risks associated with Risperdal.” In fact, by 2000, “the risks and adverse side effects associated with atypical anti-psychotics were fairly well known.”
What, then, is the basis for imposing large penalties? The court held that, to establish a SCUTPA claim, the state had to show only that Janssen’s conduct had a “tendency to deceive,” regardless whether there was any actual deception. The court dismissed as irrelevant to South Carolina the decisions of other courts finding in Janssen’s favor on claims relating to the marketing of Risperdal brought under state Medicaid fraud statutes.6
The decision contains strong language about Janssen’s alleged misconduct based on “[a]n objective review of the evidence.”7 According to the court, Janssen’s conduct was “reprehensible and in callous disregard for the health and welfare of the public” and Janssen had a “corrupt corporate culture.” The court seemed offended by Janssen’s desire to maintain market share for Risperdal and pointed to misleading efforts to distinguish Risperdal from a competitor.
The Labeling Claim
The state’s first claim was that Janssen should have made a CBE label change to strengthen the warning about the “degree of risks associated with Risperdal.” The court’s analysis is flawed,8 as it did not address whether the CBE process was available to Janssen. Janssen could not implement a CBE unless there was new and reasonable evidence of an association between Risperdal and metabolic effects,9 but the court made no finding that the medical evidence met this standard.10 The court also did not identify what language Janssen should have used, when the language should have been added or whether the FDA would have approved it.
The court acknowledged that, by January 2004, Janssen had updated the Risperdal label to include new information on diabetes and hyperglycemia, per the FDA’s September 2003 classwide directive. The court missed the fact that the CBE provision did not allow Janssen to rewrite the label the FDA had directed Janssen to use after extensive review by the FDA of the metabolic issues. The court approved penalties for use of the label in the period January 2004 to April 2007, which is after Janssen added the FDA’s required warning on metabolic effects.11
The court rejected arguments that FDA approval of the label protected Janssen from liability. Janssen argued that, under SCUTPA, there was an applicable “regulated activity” exception for conduct in compliance with state or federal law. Janssen also argued that, under the doctrine of implied preemption, a state may not impose liability that would obstruct a federal agency’s activity. The court’s analysis of these arguments is short, superficial and wrong.
Congress entrusted the approval of prescription drug labels exclusively to the FDA, leaving no room for states to impose penalties for use of a FDA-approved label. The court was misled by the state’s sophistry when it alleged liability based not on Janssen’s use of an FDA-approved label, but on its failure to implement a CBE. The court relied on Wyeth v. Levine for support.12
In rejecting implied preemption, Wyeth emphasized that Congress was aware of “the prevalence of state tort litigation” over drug labels when it declined to preempt tort claims under the Federal Food, Drug, and Cosmetic Act.13 Wyeth also noted that “Congress did not provide a federal remedy for consumers harmed by unsafe or ineffective drugs in the [FDCA]. Evidently, it determined that widely available state rights of action provided appropriate relief for injured consumers.”14 The South Carolina Supreme Court incorrectly assumed that Wyeth applied to the attorney general's action and did not recognize that actions seeking penalties for using a FDA-approved label are essentially regulatory and present fundamentally different considerations from a private-plaintiff personal injury action.15
Congress' purpose in lodging responsibility for drug labeling in the FDA was to ensure nationwide uniformity. Were the South Carolina Supreme Court’s position correct, every state could impose different labeling requirements by imposing civil penalties.
The DDL Claim
The state’s second claim was that Janssen sent a misleading DDL in November 2003 to distinguish Risperdal from Zyprexa on weight gain and metabolic effects, notwithstanding the FDA-required classwide label change that treated all drugs in the class as having equal metabolic risks. An April 2004 DDMAC letter concluded that the DDL letter was misleading. Janssen argued that the DDMAC letter was inadmissible hearsay and unfairly prejudicial. The court dismissed these evidentiary arguments in two sentences and ignored the fact that such warning letters are not definitive FDA rulings, but are informal letters of complaint.16 Moreover, a state may not punish a person for failure to comply with federal law, at least where exclusive enforcement is entrusted to a federal agency.17
The court affirmed a $28 million penalty for sending the DDL to 7,184 South Carolina doctors, and another $72 million for making follow-up visits to the doctors by sales representatives. With South Carolina representing about 1.4 percent of the U.S. population, the decision supposes that all states together could have imposed total fines of $7.2 billion based on one letter that had not been shown to deceive anyone.
A Pyrrhic Victory
Janssen’s only substantive argument accepted by the South Carolina Supreme Court was that the limitation period ran from January 2004, which was three years before the state and Janssen entered a tolling agreement. The court found that the attorney general knew or should have known of the problem with the Risperdal label before January 2007. This resulted in a nearly $200 million reduction in the total penalties. Although this victory for Janssen was significant, the court held that each use of the label during the three years prior to January 2007 was actionable on the theory that each individual violation restarts the statute of limitations as to that violation.
The result of this ruling is that, for SCUTPA claims brought by the attorney general, there is a new limitation period every day. It does not matter that the attorney general has long known about the misconduct. The attorney general may go back to penalize each instance for the last three years if he chooses to take action.
Implications
The South Carolina Supreme Court’s decision may be reviewed by the U.S. Supreme Court, as several federal issues are implicated, including excessive fines, due process, the First Amendment and conflict preemption. Assuming the decision is not reversed, what are the implications?
First, the decision may set a dangerous precedent because the court relied heavily on the apparent absence of a requirement to establish that anyone was misled or harmed to establish liability. And, although the court acknowledged that such evidence should play a role in assessing civil penalties, it nonetheless imposed enormous penalties without evidence of harm.
Second, the ruling may have some influence on other state courts. The opinion’s flaws, however, should dissuade courts from relying on it. In addition, nearly all the substantive holdings were made in the context of a failure to preserve the issues for appeal.
Third, courts should closely examine whether consumer protection statutes, such as SCUTPA, apply to prescription products. The South Carolina decision, combined with other state appellate courts decisions — which rejected the use of Medicaid fraud acts to punish Janssen — make it likely that state attorneys general will focus similar cases on consumer protection statutes. But these statutes, such as the Federal Trade Commission Act on which they are typically modeled, are aimed at consumer products and not prescription drugs and devices, where a learned intermediary has a professional duty to know the risks and benefits for each patient.18
Endnotes
1 State ex rel. Wilson v. Ortho-McNeil-Janssen Pharm. Inc., No. 27502, 2015 S.C. LEXIS 83 (S.C. Feb. 25, 2015).
2 The CBE provision, 21 C.F.R. § 314.70(c)(6)(iii), states that the manufacturer may unilaterally make label changes only to reflect “newly acquired information,” including “add[ing] or strengthen[ing] a contraindication, warning, precaution or adverse reaction for which evidence of a causal association satisfies the standard for inclusion in the labeling.”
3 DDMAC was renamed the Office of Prescription Drug Promotion in 2011.
4 The attorney general persuaded a federal judge to remand the action on the ground that “[t]he Defendants’ liability will solely depend upon their respective breach of duties as defined and created by state law.” South Carolina ex rel. McMaster v. Janssen Pharm. Inc., 2007 U.S. Dist. LEXIS 49904 (D.S.C. July 10, 2007). In fact, the labeling claim depended on whether Janssen could and should have used the CBE provision, and the DDL claim depended largely on DDMAC’s view that the DDL violated the Federal Food, Drug, and Cosmetic Act (FDCA).
5 Ortho-McNeil-Janssen Pharm. Inc. v. State, 432 S.W.3d 563, 566 (Ark. 2014).
6 See id.; Caldwell ex rel. State v. Janssen Pharm. Inc., 144 So. 3d 898 (La. 2014); Commonwealth v. Ortho-McNeil-Janssen Pharm. Inc., 52 A.3d 498 (Pa. Commw. Ct. 2012). In the Arkansas action, the court dismissed the Medicaid fraud claim, but remanded the claim under the Arkansas Deceptive Trade Practices Act based on the DDL.
7 The court did not state that its recitation of the facts was written in the light most favorable to the verdict-winner.
8 The opinion found that Janssen did not preserve and, therefore, waived a number of errors attributed to the trial court. We do not comment on this point, but, in every case where the court found waiver, it went on to reject the unpreserved argument.
9 21 C.F.R. 201.57(e) (effective until June 29, 2006).
10 The court’s erroneous reading of federal law is shown by the opinion in Marcus v. Forest Labs. Inc. (In re Celexa & Lexapro Mktg. & Sales Practices Litig.), 2015 U.S. App. LEXIS 2632 (1st Cir. Feb. 20, 2015). There, the First Circuit explained that the CBE process was unavailable because the plaintiffs’ claims were based on information “plainly known to FDA” and not “newly acquired information” for purposes of enabling a CBE label change.
11 The court held that the statute of limitations barred penalties for using the label in earlier years.
12 555 U.S. 555 (2009).
13 Id. at 575.
14 Id. at 574 (emphasis added).
15 Regulating the use of the CBE provision is “hardly a field which the States have traditionally occupied.” Buckman Co. v. Plaintiffs’ Legal Comm. 531 U.S. 341, 347 (2001) (internal quotations omitted).
16 According to the FDA manual, “A Warning Letter is informal and advisory. It communicates the agency's position on a matter, but it does not commit FDA to taking enforcement action. ... FDA does not consider Warning Letters to be final agency action on which it can be sued.”
17 See Arizona v. United States, 132 S. Ct. 2492, 2502–03 (2012) (a state may not “impose its own penalties” for federal immigration offenses because doing so would “conflict with the careful framework Congress adopted”).
18 The “Drug Amendments” to FDCA made clear that the FDA, and not the FTC, has jurisdiction over prescription drug advertising as it relates to safety and efficacy. See 21 U.S.C. § 352(n).
