What's Wrong with Antianxiety Drugs?
A possible path towards better options
Posted Aug 16, 2015
In the 1960s, major pharmaceutical companies began a concerted effort to find new and better treatments for problems related to fear and anxiety. However, new kinds of antianxiety medications have not resulted. The major classes of drugs used to treat anxiety today, Benzodiazepines (Valium and its spinoffs, such as Xanax and Klonopin) and monoamine altering drugs (antidepressants, such as imipramine, or serotonin reuptake inhibitors, like Prozac and Zoloft), have not changed for decades; new drugs have fewer or different side effects, but novel therapeutic mechanisms have not often resulted. Leading researchers recently reviewing anti-anxiety drug-discovery research have concluded the efforts have been “disappointing” and, more bluntly, “a failure.” Some drug companies indeed have raised the white flag of surrender, pulling out of psychiatric drug discovery research, including the search for new antianxiety treatments, because of the lack of new prospects for profitable drug treaments.
In Anxious: Using the Brain to Understand and Treat Fear and Anxiety (Viking, 2015) I reviewed factors that have impeded progress. Some are related to methodological/ research design issues in the animal studies used to test new drugs: for example, basing conclusions on the effects of a single dose of a drug (when some drugs require multiple days of treatment in humans), using only male subjects (when anxiety disorders are more common female than male humans), and using randomly selected animals rather than animals selected for study because they are more “anxious.” While the first two points are easily fixed (and in fact are addressed in some studies), the third point opens up a conceptual can of worms that is at the core of the problem with antianxiety drug research.
Potential antianxiety drugs are screened by assessing their effects on the behavioral and/or physiological responses of animals in challenging situations meant to mimic stressful circumstances in human daily life. Rats or mice, for example, are placed in novel or uncertain situations (an unfamiliar, unprotected area where their no place to hide), situations in which danger is present, or in which they undergo some adversity in order to obtain food, or in which they face the social challenge of a dominant intruder. Drugs that make the animals more exploratory and active in these situations (in other words, less behaviorally inhibited) are viewed as candidates for the pharmaceutical treatment of human anxiety.
The key assumption in this research is that behavioral responses reflect subjectively experienced, conscious emotional states: that animals and people are behaviorally inhibited in challenging situations because they feel fearful or anxious. Therefore, drugs that reduce behavioral inhibition in animals will make people feel less fearful or anxious. Drugs developed this way do help to some extent in some people. But no one would say that the drugs available today are magic bullets. I have an idea about what’s wrong. And it’s not so much about the research methodology underlying antianxiety drug development as it is about the conception of anxiety that has guided the research. That’s the can of worms I mentioned.
For some time now, a leading scientific view has been that emotions are self-contained entities controlled by localized, dedicated brain circuits. The classic example is the idea that the amygdala is the seat of all things “fear.” Thus, when we encounter a threat, the amygdala is activated, fear results, and this feeling causes the animal or person respond with protective behaviors (freezing, flight, fight, avoidance) and supporting physiological responses (activation of the sympathetic nervous system and the secretion of stress hormones in the body, and increased brain arousal). In Anxious, and in my previous post here (“The Amygdala Is NOT the Brain’s Fear Center”), I offered a different view: amygdala circuits are indeed responsible for the detection and response to threats and the control of behavioral and physiological responses that help cope with the situation, but the amygdala is not responsible for the feeling of fear. Fearful feelings, I proposed, are products of newly evolved cognitive circuits in the neocortex that process threats separately from, and in parallel with, the amygdala and other older subcortical circuits. Thus, while threats elicit both behavioral responses and conscious feelings of fear, these are different consequences that depend on different circuits. As described in my earlier post, because the conscious and non-conscious consequences often occur together, this correlation dupes us into thinking that they have a common origin in the brain.
A few more words are in order how on conscious feelings of fear come about. These feelings emerge from the integration of a variety of neural events in the brain, including perceptual representations of, and attention to, the threat, retrieval of long-term semantic and episodic memories related to the threat, and retrieval of fear schema. When these events come together in working memory, and are amplified by consequences of amygdala activation, such as increased brain and body arousal, they begin to coalesce in consciousness as a feeling of fear. For example, if you notice you are freezing and your heart is racing, and you see a snake at your feet, you have several features that match the fear schema stored in memory, and thus have ingredients that begin to compel the feeling of fear. But one important additional factor is the ability to know that it is “you” that all this is happening to. Only organisms that possess the concept of “I” or “me” can truly consciously experience fear. For if you do not know that it is “you” that is in harm’s way, the harm loses it impact.
The same general principle applies to anxiety, which is different from fear in that the threat is an uncertain future event rather than an immediately present source of harm. Feelings of anxiety, and the behavioral and physiological symptoms that often occur at the same time as such feelings, are products of different brain systems. Just as the amygdala is involved in behavioral and physiological responses when a threat is present, a companion subcortical region, the bed nucleus of the stria terminalis, and some older cortical areas (like the hippocampus), are involved in controlling similar responses in situations of uncertainty. Furthermore, the conscious feeling of anxiety, like the conscious feeling of fear, is a product of newer neocortical cognitive circuits that integrate diverse neural events. And while anxiety also requires the concept of “I” or “me,” in addition it also requires the ability to envision your future, and to anticipate what the consequences of the potential harm are for your future self.
From this view of fear and anxiety, it follows that drugs that alter the behavioral and physiological responses of animals in challenging and uncertain situations are not sufficient to make people feel less fearful or anxious. This is why a person with social anxiety who is on anti-anxiety medication might be more willing and able to attend parties, but still feel anxious while there.
Drug companies are giving up on antianxiety drugs because the drugs don’t have the desired effect on the way people feel. But conscious feelings of anxiety, and the worry and rumination that go with these, are only part of the problem in people who have anxiety disorders. To successfully treat problems with fear and anxiety, you not only have to change to change anxious feelings, you also have to alter the non-conscious brain systems that process present and anticipated threats and trigger behavioral and physiological responses that help cope with these. While animal studies thus have relatively little to offer in understanding and treating conscious symptoms, they have provided new clues about how to treat symptoms (such as increased threat sensitivity, hyper-arousal, and pathological avoidance) that arise from nonconscious systems. Thus, drug companies should continue their efforts, as they are well-positioned to make new discoveries about how to treat the behavioral and physiological symptoms that need to be addressed in any successful therapy.
Conscious experiences of any kind require that diverse kinds of brain events be integrated and related to representations of the self, including one’s future self. Circuits in the prefrontal cortex of the human brain are thought to be key for these kinds of brain activities. Because prefrontal circuits are more elaborate in humans than in other animals, conscious experiences, including feelings of fear and anxiety, are best researched through studies of our species.
You might be asking, don’t some drugs, like benzodiazepines, that reduce behavioral inhibition in animals also reduce anxiety itself? Yes, they do, at least to some extent. But these drugs affect behavioral inhibition by acting in subcortical (and older cortical) systems in animals and humans. In humans the drugs also affect the cognitive systems of prefrontal cortex, which, as noted, are proposed to be involved in the construction of conscious experiences of fear and anxiety. But the key issue is whether these drugs, by acting in prefrontal cortex, selectively reduce anxious feelings, as opposed to producing a more general blunting of emotional experience. Indeed, benzodiazepines impair perception, attention, vigilance, working memory,and episodic and semantic memory, cognitive processes that contribute to the construction of any kind of emotional feeling, not just feelings of fear and anxiety. As a result, although benzodiazepines may, to some extent, reduce anxious feelings in people, they may achieve this at the cost of dampening other feelings and impairing general mental acuteness.
You might also be asking, if a drug reduces a key symptom, like hyper-arousal, controlled by nonconscious systems, doesn’t that then reduce the conscious feeling of anxiety? Increased arousal does amplify anxious feelings. So a drug that reduces arousal may modulate the intensity of anxiety, but does not necessarily eliminate the feelings and worries that occupy the conscious mind of the anxious person. Failure to appreciate that different kinds of symptoms are products of different systems has led to an inadequate conception of underlies fear and anxiety disorders and how these should be treated.
Research on animals is not just useful in the development of pharmacological treatments that target nonconscious processes. It is also revealing powerful treatments based solely on behavioral conditioning procedures (more about that in a later post). At the same time, the search for better treatments for changing runaway, uncontrollable conscious feelings of fear and anxiety should proceed on both the behavioral and pharmaceutical fronts, but this is best done in studies of humans.
In sum, multiple brain systems contribute to problems with fear and anxiety. Symptoms that are manifested in subjective conscious experiences depend upon newly evolved neocortical systems, while behavioral and physiological symptoms are mediated by evolutionarily older systems that process present and anticipated threats and organize brain and body responses to cope with source of potential harm. Each system has its own unique evolutionary history and is likely to be subject to different environmental and genetic vulnerabilities, and each probably requires different treatments. This, I propose, is why the treatment of anxiety today is not more effective. Recognition of the multiplicity of the brain systems that contribute to fear and anxiety disorders is the first step towards the development of better treatments.
As always, stream the blog theme song, “I Got a Mind to Tell You” here.
The theme song of this post is “Anxious,” the title track from The Amygdaloids’ CD, Anxious, which a musical companion to Anxious (the book). From “Anxious” (the song):
Benzos, SSRIs / My my, my oh my Not perfect but OK / Help me see the way Get me through the day / At least that’s what they say
Barrett LF, Russell JA (eds.) (2014) The psychological construction of emotion. New York: Guilford Press.
Hyman SE (2013) Psychiatric Drug Discovery: Diagnosing a Crisis. Cerebrum retrieved Aug 16, 2015.
Miller G (2010) Is pharma running out of brainy ideas. Science 329: 502-504.
LeDoux JE (2014) Coming to terms with fear. Proceedings of the National Academy of Sciences of the United States of America 111:2871-2878.
Valenstein E (1999) Blaming the Brain. New York: Free Press.