Ecacy and Safety of Oral Micronized Progesterone That Converts to Allopregnanolone as an Adjunctive Treatment in Refractory Status Epilepticus: A Prospective Interventional Phase IIa Study

Background: We aimed to study the ecacy and safety of allopregnanolone which convert from oral micronized progesterone as an adjunctive treatment for Thai patients with refractory status epilepticus. Methods: Adults 18–80 years old who diagnosed with either convulsive or non-convulsive refractory status epilepticus receiving standard AED treatment, intravenous midazolam, for longer than 60 minutes were included. The intervention group received Utrogestran ® (micronized progesterone, 200 mg/capsule) via enteric feeding every 8 h for ve days. Treatments for RSE were continuing along the clinical trial period. Serum allopregnanolone levels were measured by radioimmunoassay (DetectX® Allopregnanolone Immunoassay Kit; Arbor Assays, Michigan, United States). Control patients were randomly selected from our hospital RSE cohort during 2015–2019. Continuous data were presented as the mean and standard deviation. Discrete data were assessed by number and percentage. The changes in laboratories were analyzed by paired t-test. The difference between groups were compared by independent sample t-test. The Mann-Whitney U test, Chi-square test or Fisher’s exact test were used for non-parametric parameters. A p-value < 0.05 indicated statistical signicance. All analyses were performed using SPSS version 26 software (IBM Corp., Armonk, NY). Results: The average duration of refractory status epilepticus after treatment with midazolam plus progesterone (intervention group) and without progesterone until status epilepticus termination was 25.5 and 58.4 hours, (p = 0.004), indicated that allopregnanolone signicantly shortens the refractory status epilepticus duration in vivo. Conclusions: Allopregnanolone, produced from oral micronized progesterone, demonstrates ecacy and safety in treating refractory status epilepticus as an adjunctive treatment.


Background
Several antiepileptic drugs (AEDs) are required for the treatment of refractory status epilepticus (RSE), especially when anesthetic agents are indicated as a third-line treatment for this condition. Anesthetic agents may cause signi cant side effects including hypotension, cardiac arrhythmias, prolonged intubation, long intensive care unit (ICU) stay, and increased mortality rate [1,2]. The mortality rate of adult status epilepticus (SE) in Thailand is approximately 14.5-35%. Meanwhile, the mortality rate of RSE is around 42% [3][4][5][6].
Allopregnanolone, a reduced progesterone metabolite and a positive allosteric modulator of γaminobutyric acid type A (GABA A ) receptor-mediated conduction, can reduce neuronal excitabilities and increase seizure thresholds, thus yielding a reduction of epileptiform discharges [7,8]. Previous clinical studies showed that both oral and parenteral forms of progesterone can improve seizure control as well as suppress seizure discharges in electroencephalography (EEG) [9,10]. These studies indicate that progesterone contains antiepileptic properties. Some studies using oral allopregnanolone in combination with other AEDs improved seizure control among women with catamenial epilepsies [11]. Few studies are reporting the use of intravenous allopregnanolone in pediatric and adult patients with RSE as an adjunctive treatment is bene cial [12,13]. Moreover, allopregnanolone was proposed as a safe medication. Allopregnanolone, a so-called "neurosteroid", is not available in Thailand. Nevertheless, micronized progesterone may be administered via an enteric or suppository route and be converted to allopregnanolone in vivo. Micronized progesterone is accessible in our country since it is widely prescribed by physicians for treating various obstetric and gynecological conditions. The micronized progesterone exhibits a higher bioavailability and more sustainability than other progesterone formulations and converts to allopregnanolone in vivo. We aimed to study the e cacy and safety of allopregnanolone as an adjunctive treatment for Thai patients with RSE. Also, we assessed the bioavailability of allopregnanolone that was converted from oral micronized progesterone among this patient population. Here, we determined the e cacy and safety of oral micronized progesterone for converting to allopregnanolone (an inhibitory neurosteroid) in treating Thai RSE patients as an adjunctive treatment.
In this study we use the Consolidated Standards of Reporting Trials (CONSORT) 2010 guidelines to described quality of study. The completed CONSORT checklist was described elsewhere.

Study design and patient population
This study was a Phase IIa prospective open-label interventional study that occurred from January 1st to The exclusion criteria of patients were as follows: (1) Pregnancy or lactation. (2) Serious medical problems; renal failure on dialysis, metastatic cancer, active thromboembolism, active bleeding, elevated transaminase enzymes ≥ ve times of normal value (> 160 U/L), blood pressure lower than 80/50 mmHg. (3) Absolute NPO (as the studied drug was an enteric tubal formulation). (4) Receiving drugs with potential major interaction with progesterone (e.g. estrogen, edoxaban, and venetoclax).

Study protocol
This study was a prospective interventional study. Patients in the intervention group (six patients diagnosed with RSE on intravenous midazolam infusion) received Utrogestran ® (micronized progesterone, 200 mg/capsule) via enteric feeding every 8 h for ve days ( Figure 1). Each dose was combined with 10 ml of olive oil, which acted as a vehicle to carry the agent through the nasogastric tube. Treatments for RSE, as standard guidelines using EEG as guidance, were continuing along the clinical trial period [14].
Serum allopregnanolone levels were measured by radioimmunoassay (DetectX® Allopregnanolone Immunoassay Kit; Arbor Assays, Michigan, United States). The serum allopregnanolone levels were assessed seven times for another pharmacokinetics study.
The six patients in the control group received standard treatment and midazolam infusion for their RSE and were demographically matched with the intervention group. Control patients were randomly selected from our hospital RSE cohort during 2015-2019.

Primary outcomes
Duration of the rst RSE termination, indicated by EEGs for ve days EEG SE termination based on burst suppression, background suppression, periodic epileptiform discharges with a frequency < 0.5 Hz [15] Secondary outcomes Duration of the hospital and ICU stays Mortality rate and in-hospital mortality rate compared with RSE control patients The total dose of midazolam used until the initial termination of RSE termination or until ve days elapsed Safety (adverse events) Vital signs, clinical adverse events (i.e. rashes, jaundice, hypotension) Abnormal laboratory values: complete blood count, liver function test, creatinine, coagulation test (before and ve days after progesterone administration)

Statistical analyses
Continuous data were presented as the mean, and standard deviation. Discrete data were assessed by number and percentage. The changes in laboratories for safety assessments were analyzed by paired ttest. The difference between groups (progesterone and control) were compared by independent sample ttest. The Mann-Whitney U test , Chi-square test, or Fisher's exact test were used for non-parametric parameters. A p-value < 0.05 indicated statistical signi cance. All analyses were performed using SPSS version 26 software (IBM Corp., Armonk, NY).

Patient demographics
Six patients received oral micronized progesterone (intervention group), and six patients were assigned to the control group (1:1 ratio). The mean age of patients in the intervention group was 63 years (SD 10.1, range 55-80). The most common etiology of RSE was cerebrovascular disease (2/6 patients, 33.3%). The average duration of SE before recruitment was 7.0 hours (SD 0.84). The three most commonly used AEDs during RSE were 1) levetiracetam (6/6 patients, 100%), 2) phenytoin (5/6 patients, 83%), and 3) valproate (3/6 patients, 50%). Only one patient was previously diagnosed with epilepsy (17%). Five out of six patients (83%) exhibited earlier convulsions (convulsive SE) that turned to non-convulsive SE upon joining the study. No statistically signi cant differences were found in most demographic characteristics between groups. The Status Epilepticus Severity Score (STESS) [16] was calculated. The intervention group demonstrated a relatively higher average STESS value, indicating more severe symptoms compared to the control group; however, no statistically signi cant difference was found (3.83 versus 2.67, p = 0.143). The demographic details are shown in Table 1.  Researchers showed that oral micronized progesterone was converted to allopregnanolone in vivo. The average baseline (natural endogenous) allopregnanolone level was 584.5 pmol/mL and C max at 4 h was 2,883.3 pmol/mL. At steady state, the average C trough was 2,694.4 pmol/mL, and the average C peak was 3,255.3 pmol/mL.

Clinical outcomes
The average duration of RSE after treatment with midazolam plus progesterone (intervention group) and midazolam without progesterone (control) until SE termination was 25.5 and 58.4 hours, respectively (p = 0.004, Table 2). This nding indicated that allopregnanolone signi cantly shortens the RSE duration in vivo. The average duration of the ICU and hospital stays seems shorter in the progesterone group than the control group ( Table 2). The average duration of hospital stay was 57 days for the intervention group and 58.7 days for the control group (p = 0.929). The average duration of the ICU stay was 28.7 days for the intervention group and 44.5 days for the control group (p = 0.399).
The mortality rate of the intervention group was approximately half of that of the control group (33.3% versus 66.7%, p = 0.513). The median value of total accumulated midazolam infusion until SE termination or death within ve days was 87.5 mg for the progesterone group, which was lower than 330.4 mg for the control group (p = 0.394, Table 2).
No clinical adverse events were present (i.e. rashes, anaphylactic reactions) in either group. No instances of hypotension or cardiac arrhythmias were reported in the intervention group. Meanwhile, 5/6 patients (83.3%) in the control group developed hypotension and required inotropic agents.
Changes in laboratory ndings between the baseline and day 5 after administrating progesterone were compared in Table 3. Progesterone exhibited no adverse effects on laboratory parameters.

Discussion
Allopregnanolone is one of the so-called neurosteroids. Few neurosteroids, including allopregnanolone, tetrahydrodeoxycorticosterone, and 3-alpha androstanediol, are inhibitory neurosteroids, meaning that they augment GABAergic signaling [17]. Only allopregnanolone showed in vitro and in vivo e cacy in controlling seizures in women with catamenial epilepsy as well as adult and pediatric RSE patients [9,10,13]. Our results show concordant ndings with other previous animal and human studies demonstrating that allopregnanolone improves control of seizures. Allopregnanolone showed improvement compared to the SE controls since it signi cantly shortens the duration of RSE (p = 0.004). Also, the researchers noticed that the average ICU stay, hospital stay, and mortality rates were relatively shorter than those of the control group, although no statistically signi cant differences were present. A lower midazolam dosage was needed among RSE patients using allopregnanolone compared to controls. Moreover, using progesterone treatment in RSE was quite safe, as no reports of adverse events or serious adverse events in a clinical setting or according to the laboratory results occurred. These e cacy and safety ndings were concordant with other previous international studies [13,18]. Although allopregnanolone is not commercially available in Thailand, we looked at its precursor, progesterone. Natural progesterone exhibits poor gastrointestinal absorption and a short biological half-life [19,20]. Therefore, other formulations of progesterone, including injectable, oral, and suppository micronized forms, were developed to improve and stabilize plasma levels [21].
This study was the rst clinical research study proving the role of an inhibitory neurosteroid in controlling RSE in Thai patients. We used a commercially available prodrug of allopregnanolone (Utrogestran®) in our limited resource setting. However, this study was only a Phase IIa study. Therefore, the authors suggest further studies to improve methodology, namely a multi-center, double-blind study with a longer duration and larger sample size.

Conclusions
Allopregnanolone, an inhibitory neurosteroid that is produced from oral micronized progesterone, demonstrates e cacy and safety in treating RSE as an adjunctive treatment.  Figure 1 Micronized progesterone administration t1/2 of Utrogestren®=8 hours

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. CONSORT2010Checklist.doc