Antidepressant drug development first began in the mid-twentieth century with the discovery of monoamine oxidase inhibitors and tricyclic antidepressants. Soon after, additional molecular targets and drug entities were created, eventually leading to the development of selective serotonin reuptake inhibitors. Today, antidepressants now rank among the top 10 most commonly used medications in the United States (US) and account for over $11 billion in annual sales. To help ensure the safety and efficacy of these commonly used products, in 1977 the US Food and Drug Administration (FDA) created guidelines to standardize antidepressant studies and the approval process. Although many of the recommendations outlined by FDA are still relevant, the document remains vague when describing key aspects of antidepressant trial design and much of the clinical information is outdated by today's standards. This paper will provide a general overview of the FDA-approval process, summarize FDA's position related to antidepressant trial design, and discuss the need for updates in the approval process.
HISTORY OF ANTIDEPRESSANT DRUG DEVELOPMENT
Chemical agents have been known to cause antidepressant effects since the time of the ancient Greeks when Poppy was believed to help relieve “sorrow.” This perceived effect, of course, was due to the presence of opium. “Opium cures” were even utilized well into the 19th century as a treatment of depression. Alternative agents such as dinitrile succinate, hematoporphyrin, and reserpine had begun to be used at this point in history as well.1 There is also evidence of amphetamines being utilized in the 1930s as treatment for patients with depression.2 However, it was not until the mid-twentieth century that major events in modern antidepressant drug development began.
In the early 1950s, patients with depression were largely treated with anticholinergic agents known to have psychiatric effects (e.g., chlorpromazine, thioridazine, levomepromazine).1 However, drug development took a leap with the discovery of 2 specific agents: the monoamine oxidase inhibitor (MAOI), iproniazid, and tricyclic antidepressant (TCA), imipramine. With these agents came a better understanding of the role that endogenous chemicals—such as dopamine, norepinephrine, and serotonin—play in the stabilization of mood. The 1960s brought a more widespread use of the TCAs due to the development of other agents in this class (e.g., amitriptyline, desipramine) that had fewer side effects than their predecessor. At the same time, MAOI agents began to fall out of favor due to the discovery of several adverse effects, dietary restrictions, and drug interactions.1,3,4
As psychiatric research continued into the 1970s, discussion and debate revolved around the importance of norepinephrine and serotonin (either in congruence or individually) for the successful treatment of patients with depression.3 However, by the 1980s, drug development began to shift toward serotonin as a key target in antidepressive therapy. This change may have been partially driven by the approval (1985) and quick withdrawal (1986) of the norepinephrine/dopamine uptake inhibitor, bupropion, due to seizures.5,6 Soon after in 1987, fluoxetine became the first US Food and Drug Administration (FDA)-approved selective serotonin reuptake inhibitor (SSRI).7
At this time, prescribing of antidepressant medications was becoming more common among primary care physicians. This shift has been attributed to the improved tolerability of SSRIs as compared with previously developed drugs. Antidepressant agents in this period also began to be utilized for indications outside of depression (e.g., anxiety).3 Additionally there was further development of alternative medications that affect both norepinephrine and serotonin, such as venlafaxine.1 These events likely contributed to the growing prevalence of antidepressant use in the United States, which increased by approximately 400% from 1988 to 2008.8
More recently, novel antidepressant drug approvals appear to have slowed as several recently approved medications are chemical enantiomers or different salt formulations of already approved drugs (e.g., desvenlafaxine, escitalopram, bupropion hydrobromide).7 Additionally, it has been suggested that while tolerability of these agents has improved since the 1950s, efficacy has not.1,4 Perhaps further compounding this cessation of novel drug development are outdated antidepressant approval guidelines from the FDA, which do not incentivize manufacturers to achieve greater levels of symptom improvement or target novel disease pathways. This article will serve as a primer of the FDA-approval process, review factors the FDA considers when evaluating antidepressant medications, and discuss the need for updated guidelines outlining antidepressant efficacy.
OVERVIEW OF THE DRUG APPROVAL PROCESS
The FDA is responsible for ensuring the safety and efficacy of drugs, vaccines, biologic products, dietary supplements, medical devices, foods, and cosmetics.9 Among the branches of FDA dealing with medical products, the largest is the Center for Drug Evaluation and Research (CDER), which is responsible for reviewing and approving prescription drugs in humans.
Figure 2 provides an overview of the drug approval process, which typically starts with the creation of a new molecular entity (NME). These products either contain an entirely new active moiety that has not yet been approved by FDA (also called a “new chemical entity”) or contain an active moiety that is closely related to a previously approved product (such as a different salt or ester).10,11 A drug developer must then test the pharmacologic activity of the NME in various laboratory and animal models. Once the developer is ready to test the molecule in humans, an Investigational New Drug (IND) application must be submitted to CDER. This application contains preclinical data regarding animal pharmacology and toxicology studies, manufacturing considerations (e.g., stability, production methods), and detailed protocols for the proposed clinical studies in human subjects.12
After submitting the IND application, drug developers must perform adequate, well-controlled clinical trials to prove that the new product is safe and effective. In general, these studies should be approved by an institutional review board; provide a protocol describing the study rationale, objectives, selection criteria, sample size, and methods of analysis; use a study design that permits a valid comparison with a control (e.g., placebo control, active control, dose-comparison control, historical control); employ methods to ensure the validity of the data obtained (e.g., measures to minimize bias); and provide an analysis of the results which is adequate to assess the effects of the drug.13,14
These clinical evaluations are almost always conducted in phases. Phase I studies are primarily conducted in healthy subjects and are intended to determine the pharmacokinetic effect of the drug in humans, the levels of toxicity, and the appropriate dosage range. Phase II investigations consist of small controlled trials designed to demonstrate effectiveness and relative safety in a limited number of patients. Finally, phase III trials are more extensive controlled studies intended to gather additional effectiveness data for specific indications and more precise safety information.14 Generally, the FDA requires at least 2 adequate and well-controlled studies demonstrating “substantial evidence” of efficacy in order to support approval.15 By this standard, it is important to note that the number of “negative” studies does not necessarily affect approval.
After conducting these trials and gathering enough evidence to prove the safety and efficacy of the study compound, the drug developer must submit a New Drug Application (NDA) to CDER. This NDA is the mechanism by which the drug developer “formally proposes that the FDA approve a new pharmaceutical for sale and marketing in the United States.”16 Building upon the IND, the information submitted with the NDA allows CDER to verify the drug's safety and efficacy for its proposed indication, verify the appropriateness of the proposed labeling (i.e., package insert), and assess whether the proposed manufacturing practices are adequate to ensure the strength, quality, and purity of the product.16
CONSIDERATIONS FOR THE CLINICAL EVALUATION OF ANTIDEPRESSANT DRUGS
The approval of antidepressant medications follows much of the same principles outlined above. However, FDA has created specific guidelines for the evaluation of antidepressant drugs with the intent of standardizing and clarifying the planning, monitoring, analysis, and evaluation of antidepressant studies.17 These guidelines were first published in 1977 and have not been updated or revised since that time.18 Nevertheless, the document still serves as a loose framework for antidepressant drug development in the United States.
Table 1 summarizes several requirements discussed in the FDA guidance for phase I, II, and III studies. Overall, the guidance reinforces general best practices for conducting a study, such as evaluating the new medication in a wide variety of patients and encouraging a comparison with an active or placebo control. However, the document remains vague regarding key aspects of antidepressant study design and methodology. For instance, the duration of phase II and III studies is very broad, noting that they may last days to weeks, with at least 1 study being at least 6 weeks or longer.17
Similarly, FDA does not make a specific recommendation regarding the assessment scale used to evaluate treatment effects, stating that, “Investigators are encouraged to use scales which have already been used in drug research.”17 FDA notes that experience exists with scales such as the Raskin, Beck, Wittenborn, Minnesota Multiphasic Personality Inventory, Hamilton, Zung, Overall, Spitzer, Symptom Check List, and Clyde Mood Scale, but does not make a specific recommendation as to which scale is preferred. FDA continues, vaguely stating that, “No one technique in itself is considered sufficient. Not all techniques are required,” implying that several assessment methods should be used.17 Furthermore, FDA does not offer guidance regarding the level of improvement or change in rating scale score necessary to show a benefit. There is also no mention of response or remission rates being considered as part of the efficacy analysis. These considerations have been incorporated into more recent FDA guidance documents on drug development, such as those for systemic lupus erythematosus, which define reductions in disease severity and clinical response.19
Given the age of the guidelines, it is also not surprising that much of the clinical information in the document is outdated. For example, when describing treatment comparators that should be used to test against the study compound in phase III trials, the guidance suggests TCAs and MAOIs.17 Today, these agents are generally reserved for patients who do not respond to therapy with an SSRI or serotonin norepinephrine reuptake inhibitor.20 Similarly, the diagnosis of major depressive disorder (MDD) does not reflect the current recommendations outlined in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition.21 The FDA guidance notes that defining depression is often subjective, and is not overly specific with respect to the number of symptoms or timeframe required for diagnosis (Figure 3).
The development and use of antidepressant medications has increased considerably since the 1950s, with antidepressants now ranking among the top 10 most commonly used therapeutic classes in the United States and accounting for $11 billion in annual sales.22 However, despite this wide spread use of antidepressants, novel drug development has slowed in recent years. Aside from the approval of vilazodone (2011), and vortioxetine (2013) with 5-HT1 agonist activity, the majority of antidepressants developed since 2000 are merely salt formulations or enantiomers of already approved drugs. Perhaps mimicking this trend, the FDA guidelines for approving new antidepressants has remained unchanged for the past 30 years, despite continued advancements in clinical practice. In fact, the guidelines were developed a full 10 years before the creation of SSRIs, which have now become a first-line treatment for MDD.
Although FDA has done much to regulate the development of other drug classes, such as oncology agents, the antidepressant guidelines do little to actually standardize approval requirements. Outcome measures are particularly vague, with FDA providing little guidance as to the preferred assessment scale used to measure drug efficacy. This lack of standardization has created inconsistency between antidepressant clinical trials. For example, the phase III studies used to approve desvenlafaxine were based on changes in the 17-item Hamilton Rating Scale for Depression, while the approval of escitalopram was primarily based on the Montgomery Asberg Depression Rating Scale.23,24 Furthermore, the guidance does not address the level of improvement or change in rating scale score necessary to show a benefit. Such inconsistencies have made it difficult for health care practitioners to assess the value of these medications as they enter the marketplace.
Given that FDA is charged with ensuring drug efficacy, it is important that these antidepressant guidelines remain up to date. Doing so may help stimulate the development of drugs that offer a clinically meaningful improvement in outcomes, as defined by today's standards of psychiatric practice. It may also be reasonable for FDA to require head-to-head trials evaluating patient-oriented outcomes of efficacy or safety prior to approval of “me too” products. Additionally, establishing a consensus on the number and type of depression assessment scales utilized for FDA-approval would help with standardization and the delineation of clinically useful interventions. For new agents not displaying a significant, clinically meaningful safety benefit, minimum differences in treatment effects over current therapies should be established as well.