Share This Page
The world of neuroscience and psychiatry sat up and took notice last March when the Food and Drug Administration (FDA) approved brexanolone (Zulresso) for postpartum depression. It was the first drug specifically approved for the condition, which afflicts some 15 percent of women just before or shortly after childbirth.
The event was a pivotal chapter in a neuroscience story that began three-quarters of a century ago with the 1941 discovery by Hans Selye (best known for his pioneering research into the nature of stress) that hormones including progesterone could affect the brain to induce deep anesthesia.
Fast-forward 40 years to the discovery that a number of hormones—termed “neurosteroids” by the neuroscientist/endocrinologist Étienne-Émile Baulieu, a key figure in this work—are synthesized within the nervous system itself. In their National Institutes of Mental Health (NIMH) lab, Steven Paul and colleagues showed that several of these compounds work by binding to receptors on brain cells that are activated by GABA, the most plentiful inhibitory neurotransmitter in the brain. The GABA-A receptor is the site of action of several sedating central nervous system (CNS) drugs, including benzodiazepines (Valium, Librium), barbiturates, and many anesthetics.
Neurosteroids can also bind to receptors for glutamate, the brain’s principal excitatory neurotransmitter. Paul and Robert Purdy proposed that, with its effect on both GABAergic and glutaminergic systems, neuroactive steroids (a term they coined to include synthetic analogues as well as the naturally-occurring hormones themselves) help regulate excitation throughout the brain. Excitation is a major factor in conditions such as epilepsy. Although there are many neuroactive steroids, the lion’s share of research has focused on allopregnanolone, a progesterone derivative.
Michael Rogawski, who then headed the epilepsy research division at the National Institute of Neurological Disorders and Stroke, noted with interest his colleagues’ discovery. In subsequent work at the University of California, Davis, where he is now professor of neurology and pharmacology, he developed an injectable synthetic formulation of allopregnanolone (also known as brexanolone), which was licensed for clinical development to Sage Therapeutics, a biotechnology company co-founded by Paul.
After ﬁrst investigating brexanolone as a treatment for epilepsy, Sage concentrated on depression. “A lot of parallel research comes together,” says Paul, professor of psychiatry and neurology at Washington University of St. Louis. “There’s a lot of evidence, some direct, some indirect, that GABA receptors and the inhibition system are altered and dysregulated in patients with major depression…With successful [antidepressant] treatment, levels normalize.” Although neurosteroids’ antidepressant mechanism—which circuits are involved, and how—is unknown, some research suggests that boosting inhibitory GABA activity corrects dysfunction in the hypothalamus-pituitary-adrenal axis and spurs production of hippocampal cells, which are reduced in depression.
Postpartum depression was a natural ﬁrst target. “There’s a very decent hypothesis that during pregnancy, especially the third term, levels of progesterone, the precursor of allopregnanolone, rise very high and then drop rapidly when the baby and placenta are delivered,” says Charles Zorumski, professor and head of psychiatry at Washington University (Zorumski is also a member of the Sage board of scientiﬁc advisors). “People wondered if that drop in allopregnanolone could be a trigger for postpartum depression.”
For whatever reason, the drug was a success. Two pivotal multicenter trials found that a 60-hour IV infusion of brexanolone reduced post-partum depression symptoms signiﬁcantly more than placebo. The difference became apparent within 24 hours and remained, when measured 30 days later.
This rapid effect contrasts sharply to the four-plus weeks needed for conventional antidepressants to take hold.
The approval of brexanolone is “a nice shot in the arm for the field of depression research, which went five, six decades without a new mechanism of medication,” says Eric Nestler, director of the Friedman Brain Institute at the Icahn School of Medicine at Mt. Sinai, New York. All prior antidepressants—MAOIs, SSRIs, SNRIs, and tricyclics—increase activity of one or more of the same neurotransmitters—serotonin, norepinephrine, and dopamine, he pointed out. Brexanolone, in contrast, works through its action on GABA receptors and the inhibition system.
Another new antidepressant, ketamine, which affects glutamate receptors, was approved just two weeks before brexanolone. “All of a sudden, we have two new medications that have fundamentally novel mechanisms of action,” he says. (Nestler has not been involved in neurosteroid research but has done basic science work with ketamine).
Research into allopregnanolone continues. Could it be effective for postpartum depression in an oral form, rather than the current 60-hour infusion that requires hospitalization? And might this work for major depressive disorder (MDD), a far more widespread condition? Studies in progress (both by Sage) are investigating these possibilities.
Early reports on the postpartum depression study are promising. According to a presentation at the 2019 conference of the European College of Neuropharmacology, results with zuranolone (SAGE-217)—a synthetic analogue of allopregnanolone, modified to be available when taken orally—were comparable to those with brexanolone (and significantly superior to placebo), two and four weeks after initiation of two-week therapy. Improvements were observed from the third day of treatment.
Results with SAGE-217 for MDD have been more equivocal. Although the drug was more effective than placebo in a phase II trial enrolling 89 patients with moderate to severe depression, a much larger phase III trial failed to show its superiority at 15 days, the primary endpoint. There were indications, however, that the drug may work here as well: a post-hoc analysis of the study that excluded data from non-compliant patients found a statistically significant improvement with zuranolone, as did analysis of patients with severe MDD only.
Allopregnanolone’s success against depression suggests that neurosteroids may work for related disorders, too. “In all these clinical studies, the drugs reduce anxiety as much as depression. Clinically, they are not only antidepressants but anxiolytics and sleep-promoting agents as well,” says Paul. “They promote slow wave without disrupting REM sleep.”
Studies of PH94B, a synthetic version of the pheromone androstadienol, which like allopregnanolone—and tranquilizers like Valium that stimulates GABAA receptors—suggest this neuroactive steroid may be effective for social anxiety. A phase II trial of an intranasal formulation of PH94B, developed by the biopharmaceutical company VistaGen, found that women with the disorder felt significantly less distressed during experimental public speaking and personal interaction situations when they had taken the drug 15 minutes earlier, than when they hadn’t taken it.
The treatment was effective within minutes, suggesting it would be particularly adaptable for short-term use when stressful situations arose, the study authors wrote.
PH94B has been granted fast-track status by the FDA, a designation to speed up the development and review of drugs to address unmet medical needs, and VistaGen is conducting phase III trials of the drug, according to the company.
From a broader perspective, researchers are considering therapeutic possibilities of neurosteroids for disorders ranging from schizophrenia and post-traumatic stress disorder to autism and Alzheimer’s disease. “In 2020, almost none of these things are off the table,” says Zorumski. “The reason, I think, is that these agents are affecting the two most major neurotransmitter systems, GABA and NMDA [a type of glutamate] receptors. The idea is that alterations in the balance of excitation and inhibition occur across a range of neuropsychiatric illnesses, and these compounds give you a powerful way to control excitation under conditions of stress.”
Although most of the research has focused on GABA receptors and the inhibition/excitation balance, neurosteroids appear to affect intracellular targets as well, such as processes regulating neurogenesis and mitochondria function, Zorumski says. Inflammation is believed to be a factor in a number of neuropsychiatric disorders, and there is evidence that neurosteroids have anti-inflammatory effects. One recent study identified a molecular pathway that may underlie this effect in allopregnanolone. [It should be kept in mind that the mechanistic underpinnings of neuropsychiatric disorders as a whole are imperfectly understood, and neurosteroids are unlikely to be more than part of the story.]
Researchers have explored therapeutic applications for epilepsy since the 1980s. Seizures are the result of excessive excitation in areas of the brain, which the GABA-mediated inhibition induced by these compounds would predictably counter. “There’s absolutely no question that neurosteroids stop seizures, and are prophylactic for them,” says Rogawski. But as yet, no neurosteroid-based drug has been approved for these purposes.
They would seem to have a particular role in catamenial epilepsy, a fluctuation of seizure frequency around the menstrual cycle. “Most commonly, seizures increase at menstruation,” Rogawski says: “Progesterone and estrogen rise during the second half of the cycle and drop rapidly at the time of menstruation. We demonstrated in animal models that seizure exacerbation is likely triggered by the drop in progesterone, and in corresponding levels of allopregnanolone.”
Although an allopregnanolone-based drug might be expected to address the deficiency and reduce perimenstrual seizures in women, there are only animal models and limited anecdotal evidence to support this hypothesis. “In my opinion, there is a need for a rigorous study with a neurosteroid or neurosteroid analog,” says Rogawski.
There have been no such studies as yet, he conjectured, because of technical difficulties—the natural variability in both menstrual cycles and seizure frequency would make it hard to demonstrate efficacy—and the limited potential market for this application. Negative findings in a recent NIH-sponsored trial of progesterone, the precursor of allopregnanolone, “has probably dampened enthusiasm for another trial,” he added.
Investigators have devoted considerable attention to status epilepticus, a medical emergency in which a seizure that normally last two to three minutes, persists, often despite high doses of conventional anticonvulsants. In many cases, such events can only be terminated by IV anesthetics.
“There is strong evidence neurosteroids would be useful for status epilepticus,” Rogawski says; work in his laboratory has shown endogenous allopregnanolone to be effective in animal models. Clinical trials of a synthetic allopregnanolone derivative, ganaxolone, are being conducted by Marinus Pharmaceuticals, a company that Rogawski co-founded. A small phase II trial suggested the drug could shorten seizures and obviate the need for anesthetics, the company recently reported.
Roberta Diaz Brinton, director of the Center for Innovation in Brain Science at the University of Arizona, became interested in neurosteroids in the 1980s, as a graduate student at Rockefeller University. As a postdoctoral fellow at Rockefeller, working with patients kindled an interest in Alzheimer’s disease (AD).
When she had her own lab, Brinton focused on the failure, early in AD, to encode new information, a process in which neuron regeneration is essential. Her research a few years later “brought me full circle, to the discovery that allopregnanolone promotes the generation of new neurons by stimulating neural stem cells. Where the new neurons go is the dentate gyrus of the hippocampus—which is responsible for encoding and processing new information.”
Looking further, Brinton and her colleagues found evidence that allopregnanolone also improves mitochondrial function, redressing the “bioenergetic crisis of AD. We also think it reduces the hallmark pathologies of AD—beta amyloid deposition and phosphorylation of tau.”
Neuroinflammation is another pathophysiological mechanism in AD`, and she noted that allopregnanolone “is a very effective anti-inflammatory drug.”
Given their multiple modes of action, a role for neurosteroids in other neurodegenerative disorders seems worth pursuing, she says. “Thus far, we only have preclinical evidence in a Parkinson’s model; allopregnanolone promotes regeneration of dopaminergic fibers [which are vastly depleted in Parkinson’s]. We have to pursue that more deeply. We’re focusing on common mechanisms across age-related neurodegenerative diseases; we know, for example, that there is a bioenergetic crisis in Parkinson’s, ALS (amyotrophic lateral sclerosis), and multiple sclerosis, as well as AD. There’s also evidence allopregnanolone promotes the regeneration of white matter,” the nerve fibers that connect brain regions.
AD research has gone past animal models. A phase I trial of allopregnanolone for patients with mild disease showed the drug to be safe and well-tolerated, and a multicenter phase II trial is starting up to assess its effectiveness in delaying cognitive decline and maintaining the ability to function, and to explore its effects on the brain. “Based on results of the first study, we are targeting individuals who carry the ApoE4 gene [an established risk factor for AD]. We’re beginning to apply precision medicine principles,” Brinton says.
In a separate in vitro study, Brinton’s research team is exploring the feasibility of transdermal administration of allopregnanolone for AD. While current therapies require weekly intravenous injections, this would make it possible to apply a patch—a far simpler and more patient-friendly business.
Taking an overview of neurosteroid research, Nestler sees a broad lesson. The development of allopregnanolone is “exciting… it represents a mechanism based on rational drug discovery, not serendipity.”
At the same time, he says, “the initial work on [the compound] as an antidepressant goes back three decades. Why did it take so long to get to this point? This says something about the challenges in the development of psychiatric medications.
“There are likely other mechanisms that the basic science field has talked about for years, waiting on the shelf for effective human translation, if we only have the will and resources to focus on them,” he says.