American Journal of Law & Medicine

Off-label drug use and promotion: balancing public health goals and commercial speech.

I. INTRODUCTION

Off-label promotion of prescription drugs has become a source of substantial controversy in the past decade. Before a new drug reaches the market, its safety and efficacy must be certified by the Food and Drug Administration (FDA). But the FDA does not simply approve a drug for general use. Rather, it approves drugs for the specific uses requested by manufacturers, who choose the universe of possible indications when they undertake pre-marketing clinical trials. The approval is therefore tied to a particular disease that is the subject of the manufacturer's pre-approval testing and the FDA's formal review. The conditions for which the product is approved are spelled out in the official drug label, including the dose evaluated by the FDA, and the details of administration in which the FDA has determined the drug showed efficacy. The label also describes the safety concerns related to the use.

Once a drug is approved, however, the FDA cannot control how physicians actually prescribe it. Physicians can prescribe any drug for any medical condition, even outside of the parameters of the label, for a so-called "off-label" use. Therefore, off-label use is the prescription of a pharmaceutical product at a dose and/or for a condition that the FDA has either not reviewed or not approved. Off-label uses of drugs are commonplace. For example, most drugs historically were tested and approved for use in adults; therefore, physicians who wanted to treat similar indications in pediatric patients by definition had to use the drugs off-label. (1)

The practice of off-label drug use raises important public health concerns. To continue with the previous example, differences in physiology and pharmacodynamics in pediatric patients may result in far different effects in these patients than in adults. (2) Recently, certain selective serotonin reuptake inhibitor (SSRI) antidepressants were linked, paradoxically, to increases in suicidal ideation in children. (3) More shocking was the case of valdecoxib (Bextra), a cyclooxygenase-2 inhibitor type of non-steroidal anti-inflammatory drug that, in 2001, was submitted to the FDA for approval as a treatment for pain associated with osteoarthritis, rheumatoid arthritis, menstrual cycle pain, and acute pain. (4) The FDA approved the drug for only the first three indications, rejecting acute pain as an indication because of specific dangers identified in pre-marketing trials. (5) After approval, however, the drug's manufacturer engaged in off-label promotion that successfully expanded use of the product to all pain-related indications, including the one that the FDA specifically rejected. (6) Therefore, off-label drug use can impose enhanced risk without proven benefit. (7)

In addition to individual patient risks, off-label use poses a challenge to the health care system. For example, antipsychotic drugs today are widely used off-label in elderly patients with dementia and affective disorders. (8) However, a higher mortality risk has been seen in this population with the drugs in the class, (9) including the newer atypical antipsychotic drugs. (10) In addition, the use of antipsychotics accounts for billions of dollars in annual spending by public health insurers, the lion's share of which comes from offlabel use. (11) Spending related to off-label use is of critical policy importance because it can raise drug costs for government payers like Medicaid--the federal- and state-sponsored health insurance program for the poor--that are already facing tight budgets. (12)

Current FDA regulations regarding manufacturer promotion of off-label uses weigh the rewards and risks of off-label use. Although the FDA cannot restrict physicians from prescribing drugs off-label, it does expressly forbid pharmaceutical manufacturers from promoting use of drugs for off-label purposes. While there may be some peer-reviewed evidence that an off-label use is safe and efficacious, the manufacturer has (for some reason) not submitted that evidence to the FDA for evaluation and approval. Though the Food, Drug, and Cosmetic Act (FDCA) does not expressly mention off-label promotion, the FDA considers off-label promotion to violate the provision of the FDCA that bars pharmaceutical manufacturers from introducing a new drug into interstate commerce unless it and its label have secured FDA approval. A second provision prohibits manufacturers from introducing into interstate commerce "misbranded" drugs--including drugs where the label contains misleading information, does not adequately inform safe use for approved indications, or includes information about unapproved off-label uses. Manufacturer-distributed printed and visual materials that explain the uses of the drug are considered part of the label even if not packaged with the drug. Notwithstanding these provisions, companies can respond to unsolicited questions from health care professionals and others about unapproved uses. Inquiries must be handled by the company's medical affairs office, not its sales staff, who must provide responses to the question that are narrowly tailored, balanced, and carefully documented. In January 2009, another loophole was added, as FDA allowed companies to distribute peer-reviewed scientific articles and texts discussing off-label uses, subject to several conditions. (13)

This set of policies does not sit well with pharmaceutical manufacturers. Pharmaceutical manufacturers' ability to market their products directly to physicians and patients has long been key to their ability to drive increased use of their products. (14) As a result, drug manufacturers and some affiliated activists have objected to FDA control over the scope of their marketing, arguing that it unfairly limits their ability to disseminate truthful information about their products. Numerous drug manufacturers have also actively flouted this rule. In recent years, dozens of pharmaceutical manufacturers have been involved in high-profile investigations by the Department of Justice (DO J) for improper off-label promotion, leading to billions of dollars in settlements. (15) More such settlements are on the way.

Off-label use is poised to become even more prevalent and therefore more important as a public health and economic issue. There has been an explosion in approval of biologic drugs and drugs for rare diseases, which are approved for narrow indications and therefore make the potential for off-label use more acute. (16) We may also learn more about the dangers of off-label use. (17) Physicians are starting to grow more knowledgeable about personalized factors that mediate response to drugs and reflect on the efficacy of certain off-label uses. Thus, there may be growing evidence about how off-label use of certain drugs may be inappropriate. Other trends in the health care system point to the possibility for enhanced knowledge about effective off-label uses. For example, we might learn of new evidence supporting certain off-label uses as a result of the U.S. Government's recent commitment to investing in comparative effectiveness research. (18)

Despite the growing importance of off-label drug use, the controversy over off-label promotion remains resistant to easy solutions. On one hand, the FDA's regulations seek to prevent manufacturers from encouraging unapproved uses, which might be dangerous or based on scanty evidence of efficacy. On the other hand, if an off-label use is demonstrated to be scientifically reasonable through federally-sponsored research, why should the manufacturer be restricted from promoting this outcome? (19)

In this article, I first discuss the complex intersection of the drug label and evidence-based medicine. Next, I review the phenomenon of off-label promotion and the controversy that has arisen over FDA regulations related to this field. While the goal of the drug regulatory system should be to facilitate evidence-based prescribing practices, such practices span both the approved drug label and what is currently considered off-label uses (see Figure 1). (20) The controversy over the FDA's authority to regulate off-label promotion can be traced to the essential mismatch between appropriate drug uses and the label. Thus, in the final section, I examine ways to resolve this conflict through changes to the supplemental new drug approval process that can better align restrictions on promotion with knowledge about evidence-based medicine.

[FIGURE 1 OMITTED]

II. THE BOUNDARIES OF THE PRESCRIPTION DRUG LABEL

Since the passage of the Pure Food and Drug Act in 1906, the most powerful tool available to the FDA in ensuring the health of the public through the safe and appropriate use of prescription drugs has been its power over the drug label. (21) The 1962 Kefauver-Harris Amendments for the first time gave the FDA the authority to require manufacturers to demonstrate both the efficacy and safety of a drug prior to its approval. (22) Specifically, the amendments necessitated substantial evidence of efficacy for the product's intended use that consisted of adequate and well-controlled studies. This statutory language helped lead to the modern incarnation of the drug label, and provided a clinical and legal basis for separating on-label from off-label drug use.

A. ELEMENTS OF THE DRUG LABEL

1. Efficacy

After 1962, premarketing drug development was systematized into a number of different information-collecting stages. First, a new drug is tested in the laboratory (in vitro) and in animal studies to develop basic activity and toxicology information. These steps, if successful, conclude in an Investigational New Drug application (IND), which requires a manufacturer to list "the overall plan for investigating the drug product for the following year Eincluding the] rationale for the drug or the research study [and] the indication(s) to be studied." (23) Products then undergo human testing to identify the drug's efficacy for the proposed indications. Phase 1 studies determine the maximal tolerated dose in human subjects. (24) The drug's efficacy can be first identified in Phase 2 studies, where the ability of the drug to affect the course of a disease is evaluated through controlled testing in patients with that condition. (25) Most drugs then require Phase 3 confirmatory studies, which, according to the Code of Federal Regulations, "evaluate the overall benefit-risk relationship of the drug and ... provide an adequate basis for physician labeling." (26)

Thus, the process of gathering essential efficacy information for new drug approval is highly condition-specific. The end result of successful human testing is submission of a New Drug Application (NDA) to the FDA. At the FDA, scientists review these extensive databases of studies in animals, toxicologic evaluations, and clinical trials to make a judgment about the drug. An Advisory Committee of external experts may be convened to review the data as well and provide recommendations for how the FDA should act. (27) The crux of the decision-making, however, is not simply whether the drug is efficacious or safe enough to be allowed on the market, but whether the drug's efficacy and safety justify approval for a particular intended use.

Numerous examples could be cited to elucidate this phenomenon. In 2000, the FDA approved a monoclonal antibody gemtuzumab ozogainicin (Mylotarg) as a treatment for acute myeloid leukemia (AML), a devastating cancer of the bone marrow where patients have few useful options. However, in clinical studies, gemtuzumab ozogamicin was also linked to severe and long-lasting bone marrow suppression causing low levels of white blood cells and platelets and liver problems, which could enhance mortality in AML patients receiving the treatment. (28) In making the approval decision, FDA regulators weighed the evidence of the drug's benefit against the substantial risks, and decided that the potential for efficacy justified the risk in this clinical situation. (29) Notably, what may be an acceptable risk/benefit calculus in one circumstance may not be in another. In 2004, the FDA rejected another medication, clofarabine (Clolar) as a treatment for AML, determining that its scanty benefits did not justify its substantial risks. (30)

The drug label is the summary of the evidence submitted by the companies, as reviewed by the FDA. The FDA approves a drug's label concurrently with approving the drug for marketing. As a summary of basic information and results from pre-approval studies, the label describes the particular dose (or doses) and the particular indications reviewed and approved by the FDA. The information in a drug label is presented in a systematized format delineated by regulation. (31) An "indications and usages" section contains the details of the medical conditions for which the product has been studied and has demonstrated efficacy. The "dosages and administration" section includes instructions on proper use for these conditions.

2. Safety

During the drug approval process, manufacturers also collect detailed safety information that emerges from efficacy testing. Patients' experiences with the drug under investigation are closely examined both during and after the period of administration of the drug. Side effects, ranging from minor to life-threatening, are reported to the manufacturer and compiled in the NDA application submitted to the FDA when the drug is evaluated for approval.

Thus, when a drug is approved, the drug label also includes a summary of safety information. A safety section, currently broken down into "contraindications," "warnings and precautions," and "adverse reactions," describes, in order of decreasing importance, the known side effects associated with the product. This safety information includes all existing knowledge of the drug at the time of FDA review, which includes Phase 1 and earlier studies, information derived from studies whose efficacy indications were rejected, and even information from experiences with the drug in other countries (if it had been previously approved there).

However, the breadth of the safety section is necessarily limited by the extent of the pre-approval efficacy testing, so there are important limitations to the safety information included in the official drug label. Premarketing trials involve limited numbers of patients. New drugs commonly are tested at most on a few thousand patients, so even the more common side effects may occur only in a handful of cases, and rare but deadly side effects may not emerge at all. For example, in the case of the drug nesiritide (Natrecor), approved to treat heart failure, early studies hinted at a decrease in kidney function, but none of these studies enrolled enough patients to determine the magnitude of the effect. (32) It was not until the drug was on the market for a number of years that a study combining accumulated post-marketing data with the pre-marketing data showed statistically significant and concerning rates of renal failure and mortality associated with use of the drug. (33)

Premarketing studies also are not organized to test safety. The primary and secondary clinical endpoints are almost always focused on efficacy. Therefore, premarketing studies generally lack the power to fully evaluate the frequency and importance of adverse events. (34) Inappropriate reliance on statistical testing can label rare adverse events "insignificant" if a study is under-powered to detect them. (35)

The limitations of premarket trials in evaluating safety endpoint issues can have important implications after a drug reaches the market. Erythropoietin alfa (Epogen), for example, was approved in 1989 to treat anemia associated with end-stage renal disease, a relatively rare disease. (36) After approval, however, the drug was used widely in all types of anemia, exposing millions of people to a product that was originally tested in just hundreds. Recently, data has emerged that prescribing erythropoietin alfa to patients with milder anemia increased the risk of cardiovascular events. (37)

In addition, pre-approval trials establishing efficacy may not adequately represent some classes of patients likely to receive the drug after it is approved, such as the frail elderly, or patients with other comorbidities that could interact with the drug. In these populations, the safety of the product may differ. The Randomized Aldaetone Evaluation Study (RALES) established that spironolactone substantially reduced morbidity and mortality in patients with severe heart failure. (38) While few physicians reported hyperkalemia as a side effect in the trial, physicians who used the drug in this evidence-based manner observed an abrupt increase in hyperkalemia-associated morbidity and mortality in their patients. (39) The reason was that outside of the trial, spironolaetone was prescribed to older patients with age- related renal insufficiency, potassium levels were less often monitored, and more patients received coincident ACE-inhibitors and even potassium supplements. (40)

B. OFF-LABEL DRUG USE

1. Prevalence

The drug label provides a summary of efficacy and safety based on the limited extent of pre-approval testing. But as indicated in the preceding section, because the label is based on clinical studies of specific indications, the possibility that new drugs may have efficacy in other clinical situations is high. In fact, off-label use is common, and its prevalence may be increasing. In 1985, Strom et al. studied the 1,000 most common uses of prescription drugs and found that thirty-one were for non-FDA-approved indications. (41) In 2003, using physician-reported prescribing data, the reasons for using forty-five drugs (the three top drugs in each of the fifteen leading drug classes) were analyzed for 1988-89 and 2002-03. Off-label use rates were 22%. (42) A recent study used 2001 data to define prescribing patterns for 160 drugs, including the top 100 most prescribed office-based drugs, for a total of over 6,000 drug- diagnosis pairs. That study found a 21% off-label prescription rate generally, and concluded that "[a]mong off-label mentions, most (73%) lacked evidence of clinical efficacy, and less than one-third (27%) were supported by strong scientific evidence." (43)

Off-label use is particularly prevalent in two disease categories: oncology and neuropsychiatric drugs. While few studies have detailed the patterns of use of oncology drugs for off-label indications, intermittent reports suggest this practice is frequent. In 1991, the General Accounting Office found that approximately one-third of chemotherapy treatments were off-label. (44) More recently, the American Society of Clinical Oncology reported an off-label use rate of approximately 50%. …

Log in to your account to read this article – and millions more.