Abstract
Neurons metabolize cholesterol and cholesterol-derived steroid molecules to produce a class of molecules known as neurosteroids. The important role of these molecules in modulating neuronal function is increasingly appreciated, as is their potential as therapeutics (Zorumski and others 2025). The US Food and Drug Administration (FDA) approved brexanolone, an intravenous form of the endogenous neurosteroid allopregnanolone, for the treatment of postpartum depression in 2019, and zuranolone, a similar orally bioavailable molecule, in 2023. This opens the door for the development of other naturally occurring and synthetic neuroactive steroids for the treatment of depression and a range of other neuropsychiatric conditions. This is an exciting direction of ongoing work.
Allopregnanolone (“allo”) is the canonical neurosteroid. Allo is a potent positive allosteric modulator of most GABA-A receptors, increasing their open time and, when activated by GABA, the amount of chloride they transport, thereby increasing their inhibitory activity. It is produced from progesterone via a two-step enzymatic process. Its levels rise dramatically during pregnancy and then fall precipitously after delivery. This rapid drop is thought to contribute to the development of postpartum depression, motivating the use of exogenous allo in the treatment of this condition (Paul and others 2020). Allo also increases markedly in animals after acute stress, but it decreases following chronic stress. Since allo has an anxiolytic effect, acute elevation may represent a homeostatic response, while chronic reduction is likely to contribute to the pathological effects of chronic stress, including in posttraumatic stress disorder (Pinna and others 2025).
Another class of molecules, the benzodiazepines, also acts through positive allosteric modulation of the GABA-A receptor and has well-documented anxiolytic effects, but benzodiazepines are not generally found to be effective antidepressants. Why are allopregnanolone and other neurosteroids different? It may be that mechanisms beyond modulation of the GABA-A receptor are at play. For example, allo and other neurosteroids can modulate neuroinflammation through interactions with TLR-4 (Morrow and others 2025). Allo can also modulate low-voltage-activated calcium channels and possibly even NMDA glutamate receptors (Zorumski and others 2025). More work is needed to clarify the contribution of these various mechanisms to antidepressant and anxiolytic effects, as well as to side effects, to guide future development of the next generation of neurosteroid modulators.
It also remains to be seen how broadly this set of mechanisms will be of benefit, beyond postpartum depression. The FDA recently denied a request for approval for zuranolone for the treatment of major depressive disorder, beyond the postpartum context, and a recent Cochrane review found mixed evidence for the benefit of both brexanolone and zuranolone, with significant concern for limiting side effects (Wilson and others 2025). Clearly, more work is needed to develop new agents, refine our understanding of how best to deploy them, and conduct ongoing rigorous testing in clinical populations. But It looks like therapeutic neurosteroids . . . are here to stay, as one of the few pharmacological strategies for the treatment of depression grounded in an understanding of the underlying pathophysiology.