Update on Corticotropin-Releasing Factor Pharmacotherapy for Psychiatric Disorders: A Revisionist View

The identification of corticotropin-releasing factor (CRF) in 1981 was followed by the discovery of three CRF paralogs (urocortins 1, 2, and 3) and two CRF/urocortin receptors (CRF₁, CRF₂; Bale and Vale, 2004). Because preclinical studies showed that CRF₁ receptors mediate endocrine, behavioral, and autonomic responses to stress, the pharmaceutical industry developed blood–brain barrier-penetrating CRF₁ receptor antagonists. We and others previously surveyed the pharmacology of non-peptide CRF₁ receptor antagonists and the therapeutic rationale of CRF₁ antagonists for major depression, anxiety disorders, and addiction (see Koob and Zorrilla, 2010; Zorrilla and Koob, 2010, for references). Yet, CRF₁ antagonists have still not yielded positive Phase III clinical trials, prompting the current revisionist view of the neurotherapeutic potential of CRF₁ antagonists. Our hypothesis is that CRF antagonists may be valuable in specific psychiatric disorders in which stress is a dynamic rather than chronic condition. More explicitly, we suggest that CRF₁ antagonists in psychiatry may particularly be useful in post-traumatic stress disorder (PTSD), panic disorder, and addiction.

Non-peptide CRF₁ antagonists consistently produce anxiolytic-like effects in certain animal models, such as conditioned freezing, defensive burying, acoustic startle responding, the open field, the elevated plus maze, the light–dark box, the defensive withdrawal test, and the social interaction test. A CRF₁ antagonist (R317573/JNJ19567470/CRA5626) also recently showed activity in rodent (Shekhar et al, 2011) and human (Bailey et al, 2011) panic models. These models reflect a dynamic, active response to an acute stressor and, from a face validity perspective, may reflect more the symptoms of specific subtypes of anxiety disorders rather than of generalized anxiety disorder. Indeed, CRF₁ antagonists exhibited weak activity in punished drinking and punished crossing conflict models, unlike γ-aminobutyric acid anxiolytics. Despite initial positive results, small-molecule CRF₁ antagonists have not consistently shown efficacy in animal models of antidepressant activity (Zorrilla and Koob, 2010).

CRF₁ antagonists also reduce the activation of brain stress systems in models of addiction, supporting the therapeutic potential of CRF₁ antagonists for drug dependence. Hypothalamic-pituitary adrenl-axis and extrahypothalamic CRF systems are activated during acute withdrawal from all major substances of abuse in animals. CRF antagonists blocked anxiogenic-like responses to withdrawal from cocaine, alcohol, nicotine, cannabinoids, and palatable food and blocked the development of or reduced already escalated drug self-administration in addiction models (for details and references, see Koob and Zorrilla, 2010; Boyson et al, 2011). CRF₁ antagonists also blocked stress-induced reinstatement of heroin-, cocaine-, nicotine-, alcohol-, and palatable food-seeking behavior and stress-induced reactivation of conditioned place preference for opioids and cocaine (Koob and Zorrilla, 2010).

No CRF₁ antagonist has successfully completed a Phase III trial. R121919 and PF-00572778 were abandoned due to liver enzyme elevations (NCT00580190). The development of ONO-2333 Ms (NCT00514865) and CP-316,311 were halted because of negative efficacy in double-blind, placebo-controlled trials for major depression (Zorrilla and Koob, 2010). Verucerfont (GSK561679) also lacked efficacy in a major depression trial (Protocol # CRS106139). Pexacerfont (BMS-562086) was ineffective against generalized anxiety disorder (Coric et al, 2010). Trials of verucerfont and emicerfont for social anxiety disorder have been completed with undisclosed results (NCT00555139). Relevant to the hypothesis proposed herein, GlaxoSmithKline and NIH are currently evaluating verucerfont against startle in healthy women (NCT01059227), in women with PTSD (NCT01018992), and against stress-induced alcohol craving in anxious women (NCT01187511). A trial for pexacerfont has likewise been initiated in anxious alcoholics by Bristol Myers Squibb and NIAAA (NCT01227980). Several other candidates are earlier in the pipeline, or their status has not been publicly updated by the pharmaceutical industry (eg, GSK586529 [NCT01059227], SSR125543 [NCT01034995], antalarmin). Should results from these trials concur that CRF₁ antagonists are ineffective for chronic anxiety and depression, a re-evaluation should be considered with emphasis on certain anxiety disorders, such as PTSD and possibly panic disorder, and on addiction disorders.