Brexanolone for the treatment of patients with postpartum depression

Summary

On March 19, 2019, the United States Food and Drug Administration (FDA) approved Zulresso (brexanolone) for intravenous use for the treatment of postpartum depression (PPD) in adult women. The decision was based on three recent clinical trials following an FDA priority review and breakthrough therapy designation. Brexanolone is now available through a restricted process called the Zulresso Risk Evaluation and Mitigation Strategy Program that requires the drug to be administered by a healthcare provider in a certified healthcare facility. Brexanolone represents an important new treatment option to address treatment-resistant depressive symptoms. In this article, we discuss the current critical need for PPD treatments, the mechanisms of brexanolone action, and the efficacy and drug safety studies that led to FDA approval. Additionally, we discuss some limitations of the current formulation, specific populations of women that might benefit from this treatment, and how new drugs on the horizon may increase the ability to treat PPD in a variety of patient populations.

Background

Mood disorders represent a class of emotional disturbances characterized by a loss of interest in previously enjoyable activities (anhedonia) and/or extreme sadness, and are one of the leading causes of disability in the world (1, 2). Women are twice as likely as men to suffer from mood disorders such as major depressive disorder (3). The perinatal period—before and after birth—is a time of increased risk for women to present with mood disorders (4).

Postpartum depression (PPD) is one of the most common complications of childbearing, with several independent estimates showing that about 11-19% of women suffer from postpartum depressive symptoms in the 3 months following birth (5–7). PPD has a serious and widespread negative impact on mothers, babies and families. Data from over 5,000 families shows that PPD was present in a significant proportion of mothers and was associated with undesirable maternal behaviors, including unhealthy feeding and sleep practices, and fewer positive parent-infant interactions (8).

A recent meta-analysis of 122 PPD studies revealed an association of PPD with outcomes such as negative maternal physical and psychological health, poor social relationships with partners, and difficulty in bonding with the infant (9). Infants of mothers with PPD had lower motor development scores at 6 months of age and lower language scores at 12 months of age than infants of nondepressed mothers, suggesting that maternal mood has a serious impact on important developmental milestones in the infant (10). These negative outcomes of PPD on the health of mothers, babies and their families are present worldwide, indicating that there is an underlying biological etiology of PPD (11–13).

While the consequences of PPD are well recognized, until now there has been little progress in the development of pharmacological treatments specifically for this unique population. Current pharmacological treatment for PPD is the same as for depressive disorders in nonpregnant adults, namely antidepressant drugs. There are many disadvantages of current first-choice antidepressants, which include selective serotonin reuptake inhibitors (SSRIs), serotonin and norepinephrine reuptake inhibitors (SNRIs), and atypical antidepressants such as bupropion. Each of these classes of drugs can require weeks to months to exert their antidepressant effects, leaving patients suffering for a significant amount of time following the start of treatment. For postpartum women and their infants, this delayed effectiveness can promote irreparable harm to the mother-infant bond and to the development of the infant in what is clearly a critical period. Additionally, there is a barrier to taking these drugs during and following pregnancy due to the passage of SSRIs through the placenta and into breastmilk. SSRIs are considered safe to use in pregnant and lactating women, but little is known about their long-term effects on the development of the infant, including their effects on the neonatal brain where serotonin is critical to neurodevelopment (14–17). This risk needs to be balanced with the often serious consequences of untreated depression in the mother. As such, clinicians have been seeking alternatives that are faster-acting and have lower risk profiles for infants.

In a major breakthrough, the first drug ever developed specifically for PPD was recently approved. Sage Therapeutics received U.S. Food and Drug Administration (FDA) approval for Zulresso (brexanolone), an intravenous formulation of allopregnanolone (18).

Pharmacology

Brexanolone is a synthetic proprietary intravenous formulation of the endogenous neurosteroid allopregnanolone and is chemically identical to allopregnanolone (19) (Fig. 1). Although it is a steroid molecule, allopregnanolone acts rapidly via nongenomic mechanisms to modulate neurotransmitter receptor function (20). Allopregnanolone is a potent positive allosteric modulator of synaptic and extrasynaptic GABA_A receptors (GABA_ARs), prolonging the opening time of chloride channels within GABA_ARs, which amplifies inhibitory neurotransmission (21). Allopregnanolone can also directly activate the receptor at micromolar concentrations (22–24). Allopregnanolone thus exerts similar effects to those of other GABA_AR positive allosteric modulators, such as benzodiazepines, including their sedative effects (25–29), although there are subtle differences related to their unique binding sites, potency and selectivity for GABA receptor subtypes (30). It can therefore be expected that elevated allopregnanolone levels will enhance inhibitory synaptic function, as well as extrasynaptic inhibition, and thereby reduce overall neuronal activity (31). It remains unknown where in the brain the critical circuits underlying both the genesis and the treatment of PPD are located, and to what degree these circuits overlap.

Figure 1.

Chemical structure of brexanolone.

Understanding the underlying biological mechanisms of PPD is an area of active investigation. There are several possibilities, including a hypersensitivity to the normal dynamic hormonal changes during pregnancy and postpartum, dysregulation of the hypothalamic–pituitary–adrenal (HPA) stress axis during pregnancy and postpartum, and genetic mechanisms such as single nucleotide polymorphisms in the serotonin transporter and estrogen receptor α genes (32–35). While the exact mechanisms are unclear, researchers generally agree that the dynamic changes in hormones during and following pregnancy are key in promoting the onset of PPD. The stress hormone cortisol and the ovarian hormones progesterone and estradiol all rise 3- to 10-fold during pregnancy, as do their metabolites, and then precipitously drop at parturition (36, 37). One neuroactive metabolite of progesterone is allopregnanolone. Consistent with the rise in progesterone during pregnancy, levels of allopregnanolone increase throughout gestation and peak at term, with levels increasing from 4 nmol/L at 8-10 weeks after conception to > 40 nmol/L at 36-38 weeks, and then plummeting back to 1 nmol/L at 4-6 weeks postpartum (31, 38, 39).

Because it is an allosteric modulator of GABA receptors, most theories about the relationship between allopregnanolone and PPD focus on GABAergic inhibition (40). Abnormally low levels of allopregnanolone during pregnancy are associated with increased depressive symptoms and increased negative emotional responses (41–43). Similarly, polymorphisms in genes for steroid hormone processing enzymes that decrease allopregnanolone production are associated with increased PPD risk (41). One hypothesis to explain PPD is that there are insufficient levels of allopregnanolone during pregnancy, leaving the inhibitory system too weak to cope with the elevated cortisol levels during pregnancy, the fall in allopregnanolone levels after birth, or the stresses associated with being a postpartum mother, including dysregulated sleep.

Alternatively, the naturally elevated levels of allopregnanolone during pregnancy may promote a homeostatic downregulation of GABA_ARs and inhibitory synaptic function, thus leaving the brain without sufficient inhibitory tone in potentially critical synapses (44). Women at risk for PPD also have lower blood plasma GABA concentration during the postpartum period, and lower GABA levels are correlated with risk for PPD (45). Furthermore, it has been hypothesized that when allopregnanolone levels fall sharply at parturition, GABA_AR numbers and inhibitory synaptic function recover from their downregulated state too slowly in some women, resulting in weakened inhibition and thereby triggering the onset of PPD (43, 46). Deficits in GABAergic function would also contribute to depressive episodes by reducing sensitivity to allopregnanolone and preventing its endogenous anxiolytic, sedating and prosomnolence effects during and after pregnancy (47).

Pharmacological studies show that the subunit composition and activated phosphorylation state of GABA_ARs influence the sensitivity of the receptor to allopregnanolone (48, 49). The δ subunit of GABA_ARs is of particular interest to allopregnanolone and brexanolone actions because its incorporation into the GABA_ARs that are located primarily at extrasynaptic sites where they respond to tonic levels of GABA increases the ability of allopregnanolone to potentiate GABA responses. Relative to other conformations of the GABA_ARs, those with the δ subunit require lower concentrations of allopregnanolone to reach EC₅₀, or the half maximal response (48). Expression of GABA_AR δ subunit decreases during pregnancy and postpartum, which can be predicted to limit the ability of allopregnanolone to potentiate inhibition. In mice, deletion of the GABA_AR δ subunit results in abnormal postpartum maternal behaviors (46, 50). Brexanolone treatment of mice improved these abnormal postpartum behaviors in this PPD-relevant preclinical model (51). These findings suggest that the composition and quantity of GABA_ARs in the brain may determine how women respond to the dynamic hormonal changes of pregnancy and postpartum, leading to increased risk for PPD.

Regardless of mechanism, the shift in GABA_AR sensitivity leaves the pregnant brain in a state of decreased GABAergic tone immediately prior to the most dramatic change in allopregnanolone levels, the sudden drop-off following parturition. Provision of supplemental allopregnanolone in the form of brexanolone can be predicted to counteract this change and restore normal function. As the action is thought to be directly on GABA_ARs in the cell membrane, the effects of supplementing allopregnanolone levels with brexanolone are expected to occur rapidly. This allows clinicians to provide faster relief of symptoms upon diagnosis of PPD than has been possible with traditional antidepressants. Although these direct effects on GABA_ARs likely lead to the rapid effectiveness of brexanolone, there may be other synaptic mechanisms underlying the prolonged relief of symptoms seen with this treatment, including delayed but sustained changes in GABA_AR expression or function. Furthermore, allopregnanolone can be metabolized to 5α-dihydroprogesterone (5α-DHP), which binds with high affinity to intracellular progesterone receptors that regulate transcription and may underlie sustained functional changes (52).

A final open question is the role of sedation and regularization of sleep cycles in the antidepressive response to brexanolone. Because it is a positive allosteric modulator of GABA_ARs, allopregnanolone promotes sleep in a pattern similar to that seen with other agonistic modulators of GABA_ARs (29). As any parent of a newborn will know, sleep disruption is a common consequence of childbirth. Patients administered brexanolone in an inpatient setting report that they experienced better sleep (53). Less fragmented sleep as well as emotional support and help with infant care from caregivers in the inpatient setting may also contribute to the considerable placebo response observed in the brexanolone clinical trials.

Clinical Studies

Following the required clinical pharmacological studies in healthy volunteers to determine metabolism, abuse potential, safety, bioavailability and breast milk concentrations of brexanolone, clinical studies were undertaken. Pharmacokinetic studies revealed that brexanolone has a biphasic elimination, with an initial half-life of 40 minutes, resulting in rapid clearance of the drug.

Three double-blind, randomized, placebo-controlled trials were conducted at four clinical research centers (study 1) and subsequently 30 clinical research centers and specialized psychiatric units in the United States (studies 2 and 3) (53, 54). Brexanolone was administered in an inpatient setting with 60 hours of intravenous infusion at 60 μg/kg/h and 90 μg/kg/h doses and compared to placebo. To enhance tolerability, the dose was stepped up and then back down gradually over the 60 hours (30 μg/kg for 0-4 hours, 60 μg/kg for 4-24 hours, 60 μg/kg or 90 μg/kg for 24-52 hours, 60 μg/kg for 52-56 hours, and 30 μg/kg for 56-60 hours). An important factor to consider is that placebo treatment involved a similar 60-hour placebo injection with associated inpatient clinical care. Women included in the studies had ages ranging from 18 to 45 years, were ≤ 6 months postpartum, had a major depressive episode sometime during the third trimester of pregnancy or up to 4 weeks postpartum, were displaying some depression symptoms according to the 17-item Hamilton Rating Scale for Depression (HAM-D) score, and were willing to temporarily cease breastfeeding. Brexanolone was the primary treatment for the patient’s PPD, as 71% of subjects did not receive additional antidepressants during the clinical trials.

The primary endpoint of the clinical trials was overall change in the 17-item HAM-D score at the end of the 60-hour infusion compared to baseline (start of infusion). HAM-D is one of the most widely used measures of depression severity (55). HAM-D responses were also assessed at 72 hours, 7 days and 30 days following the start of infusion. The secondary endpoint of the clinical studies was change from baseline in HAM-D total score at day 30. In all three independent clinical studies that were completed, a statistically significant improvement in HAM-D scores, compared with baseline scores, was achieved. The change in least square (LS) mean HAM-D scores ranged from 17.7 to 19.5 in response to brexanolone, although the response was less clearly different from placebo when measured with other depression scales. It must be noted that HAM-D includes questions related to sleep, which could have contributed to the lower score. Furthermore, placebo infusions in the inpatient setting also caused a reduction in LS mean HAM-D scores of 14.0 in these trials (53, 54).

Safety

Brexanolone targets GABAergic synapses and extrasynaptic GABA_ARs nonspecifically throughout the brain, leading to concerns about adverse sedation-related events. Indeed, sedation-related events, including dizziness, somnolence, sedation, fatigue and loss of consciousness, occurred at a higher rate in brexanolone-treated patients (27%; 38/140) than in placebo-treated patients (14%; 15/107) in the first 24 hours of the 60-hour infusion (53, 54). In studies 2 and 3, 5/130 (4%) patients had excessive sedation following brexanolone treatment that was considered a severe adverse event leading the patients to withdraw from the study (54). Due to these safety concerns, the approval of Zulresso was delayed by 3 months while the FDA implemented risk evaluation and mitigation strategies (REMS) to minimize the risk for loss of consciousness. REMS is an established mechanism for the FDA to ensure favorable outcomes and has been used successfully in the past. The key components of REMS with Elements to Assure Safe Use include enrollment of the prescriber in REMS, enrollment of all patients in a registry to collect data and further characterize any loss of consciousness, and restrictions on the drug’s distribution to certified healthcare settings. Observation of patients by a healthcare provider in a certified facility allows for the confirmation of the discontinuation of the sedation events following treatment. Importantly, any excessive sedation observed in brexanolone patients was reversed within 15 minutes of stopping the infusion.

There were few other adverse effects reported in the clinical trials. Importantly, there was no evidence of increased suicidality (53, 54). While women enrolled in the clinical trials were required to cease breastfeeding for the duration of the infusion, a lactation study demonstrated a low Relative Infant Dose (RID) of 1.3% at maximum, which is well below the recommended threshold of < 10% RID (56). Ultimately, it was recommended that breastfeeding during infusion be discussed by the mother and physician.

Conclusions

PPD is a significant health issue for new mothers, and the lack of effective treatment is a considerable issue for maternal mental health. It is exciting that the FDA has approved Zulresso (brexanolone), the first antidepressant developed specifically for PPD. The mechanism underlying the rationale for brexanolone treatment for PPD is based on a strong scientific foundation and is supported by numerous preclinical and clinical studies. Amid this excitement, there are issues that need to be addressed, including better identification of the target population, further refinement of the pharmacology and lowering the barriers to access. Increased recognition of the risks of PPD and screening for symptoms in new mothers are needed to improve current low rates of diagnosis and treatment (57).

Major depression, including PPD, is a heterogeneous disorder that likely has more than one biological etiology. Now that brexanolone has been approved, clinical experience will hopefully reveal which patients are likely to respond positively to this treatment. Adverse childhood experiences (ACEs), which encompass all types of abuse, neglect and other potentially traumatic experiences that occur prior to the age of 18, predict a variety of negative health outcomes, especially for women during periods in the lifespan in which there are dynamic periods of hormonal changes, including for PPD (58–60). Studies examining the impact of ACEs on maternal outcomes in pregnancy in mice and humans showed the same negative correlation between maternal ACE history and HPA axis stress reactivity, suggesting that careful consideration of patient psychiatric history is important in determining the best treatment options (61). Other relevant factors that should be considered include maternal HPA axis reactivity and measures of sleep disturbance and anxiety.

Due to high intrinsic clearance and low volume of distribution, brexanolone is ideally suited for parenteral administration with a fast-on/fast-off kinetic profile (62). Practically, the requirement for intravenous administration in an inpatient setting is a major barrier to widespread use, both financially and socially. The requirement of a monitored 60-hour inpatient hospitalization and fees combined with a drug price suggested at around USD 34,000 per treatment have raised concerns among the public. To this end, Sage Therapeutics has another compound, SAGE-217 (zuranolone), in development. This compound differs from brexanolone in the addition of a methyl group at C3 and a pyrazole substitution at C-21, which together confer oral bioavailability to the compound while maintaining modulatory efficacy at the GABA_ARs. Sage also states that SAGE-217 has low intrinsic clearance with potential for once-daily dosing (63). Although the phase III clinical trial results have yet to be published, Sage has announced that SAGE-217 provided a significant improvement in the HAM-D score compared with placebo (64). This effectively means that SAGE-217 could provide the therapeutic benefits seen in brexanolone, but in a once-daily oral dose that should significantly drop the overall cost of treatment. Short-course oral treatment with SAGE-217 in an outpatient setting will hopefully reduce this potential barrier to access.

In summary, the approval of brexanolone for the treatment of PPD is a positive step forward for the treatment of women’s mental health. While there are currently some barriers to access, rapid-acting treatments such as Zulresso are likely to have a significant effect on the long-term health of the mother, baby and family.