Effectiveness of ACTH in Patients with Infantile Spasms

(1) Background: West syndrome is a severe, refractory, epileptic syndrome that usually appears in infancy or early childhood. ACTH is one of the more effective drugs for treating this condition. (2) Aim of the study and methods: The objective of our study was to examine short-term efficacy (during treatment schedule) and long-term outcome of intramuscular 0.02 mg/kg/day ACTH (tetracosactide) depot, used concomitantly with other antiepileptic drugs (AEDs) in patients with infantile spasms who did not achieve seizure cessation or relapse when taking only the AEDs. The drug efficacy was evaluated in retrospective and prospective analyses of 50 patients diagnosed with infantile spasms. (3) Results: Complete cessation of spasms was achieved in 42 cases (84%). EEG improvement was seen in 41 (82%) patients who responded to ACTH therapy. Information on the clinical course of 28 patients was obtained duringlong-term follow-up. In 17 (60.7%) cases, seizures were still present. Normal or near-normal development was observed in 11 out of 28 children (39%). ACTH used concomitantly with other AEDis a highly effective treatment with acceptable side effects. (4) Conclusion: Randomized controlled clinical trialswith long-term follow-up are needed to compare the effectiveness of ACTH in polytherapy and monotherapy. Dyskinesias as a potential side effect observed in our study group should be investigated in the following studies.


Introduction
Infantile spasms (IS) is a challenging epileptic syndrome to treat. Its incidence is approximately from 1:4000 to 1:6000 live births [1], constituting [13][14].5% of all cases of infantile epilepsies [2]. It can also be called infantile epileptic encephalopathy since the continuous, disorganized, and arrhythmic electrical activity of neuronal networks may influence brain maturation, cause arrest or regression of previously gained functions, or make infant milestones impossible to achieve [2]. Infantile spasms consist of three features: epileptic spasms, developmental arrest or regression, and hypsarrhythmia-an interictal EEG pattern. However, not all of them must be present. Hypsarrhythmia together with epileptic spasms in clusters are obligatory features to diagnose West syndrome (WS), according to the West Delphi Proposal, but some studies use the terms West syndrome and infantile spasms as synonyms. Neurodevelopmental arrest or regression before the onset of spasms is not required for the diagnosis of WS by the aforementioned Proposal [3]. This epileptic disorder occurs most frequently within 12 months of life, with a peak between 4 and 7 months. However, cases with later onset have been reported [4].
Epileptic spasms are brief contractions (flexions, extensions, or mixed) of the neck, trunk, and upper and lower limbs or head nodding, occurring in clusters, frequently on

Patient's Characteristics
Fifty children (29 male, 21 female) diagnosed with infantile spasms were included in the study. The age at the onset of epileptic spasms varied between 1 and 11 months, while the mean was 5.2 ± 2.2 months. The age at the onset of ACTH therapy was 8.6 (6-9) months. According to the ILAE classification, 41 (82%) patients could be classified as symptomatic. A detailed presentation of patients with known etiology is shown in Table 1. Other types of seizures prior to the onset of epileptic spasms were present in 17 (34%) patients, and developmental delay was observed during hospitalization in 45 children (90%, the first developmental evaluations were performed at onset in hypsarrhythmic state), microcephaly in 13 (26%), and dysmorphic features in 17 (34%). The family history was relevant in the case of 10 (20%) patients. Ten (20%) patients were born preterm. Abnormalities in MRI were present in 35 cases (70%).

Study Pattern
This study was conducted by a retrospective and prospective analysis of patients hospitalized in the Department of Pediatric Neurology of the Medical University of Silesia in Katowice diagnosed with infantile spasms who were admitted between March 2009 and September 2017. Table 2 shows the onset of epilepsy depending on presence or absence of hypsarrhythmia. Patients who were not treated with ACTH were excluded from the study.
ACTH was not the first-line treatment in this population and was used concomitantly with other antiepileptic drugs. AEDs taken by the participants of the study during and before the onset of tetracosactide therapy were noted. Tetracosactide was used at a dose of 0.02 mg/kg/d for two weeks and tapered to zero over the next 4 or 6 weeks-the exact treatment schedule is presented in Section 3 ( Table 3). Patients with treatment lasting 6 weeks had the same schedule without the last 2 weeks with one dose per week. The time of ACTH administration was shortened to 5 weeks in one patient because she was qualified for the neurosurgical operation of hypothalamic hamartoma. Video EEG was taken in all patients. Video EEG and EEG were performed at two-week intervals and after the end of ACTH treatment. Improvement of EEG recording in patients with hypsarrhythmia meant its withdrawal and, in other cases, reduction of paroxysmal changes and their amplitude and the dominance of slow theta and delta waves. Table 3. Schedule of treatment shows frequency of 0.02 mg/kg/d ACTH administration in particular weeks of treatment.

Weeks of Treatment
Frequency of Administration As far as the causes of epilepsy are concerned, patients were categorized according to the 2017 ILAE classification as structural, genetic, infectious, metabolic, immune, and unknown. Some patients could be classified into more than one category. The primary clinical outcome was cessation of spasms, defined as the lack of observed seizures until the 14th day of ACTH treatment, and a seizure-free interval during hospitalization. Patients who achieved cessation of epileptic spasms did not have seizures for 28 consecutive days or more from the last witnessed seizure.

Statistical Analyses
Continuous variables were expressed as the mean ± standard deviation or median (lower quartile-upper quartile), when distribution was not normal. Shapiro-Wilk test and Q-Q plots were used to test normality. Categorical variables were presented as numbers and percentages using Open Office (Apache Open Office 4.1.2 Copyright 2015 The Apache Software Foundation). Continuous data were compared by means of the Mann-Whitney U test. To compare the categorical variables in the short-and long-term outcome, the chisquare test with Yates's correction and Fisher's exact probability test were used, respectively. The data were compared using Statistica 12 (Tulusa, OK, USA). Statistical significance was defined as p < 0.05.

Treatment Schedule
All patients took at least one AED prior to tetracosactide, and they did not achieve cessation of spasms or had recurrence of epileptic spasms when they started ACTH therapy. The time from diagnosis of epilepsy to administration of the first AED was longer than 30 days in 7 symptomatic patients. The average time lag between the onset of epileptic spasms and the beginning of ACTH therapy was 1.5 (1-3) months. ACTH was used with many other AEDs as add-on therapy. The median number of drugs used prior to tetracosactide was 3.66 ± 1.85 (Table 4). Some patients were undergoing discontinuation of other AEDs when they started ACTH therapy. The following drugs were used during the first two weeks of ACTH therapy:valproate in 41 (82%) patients, vigabatrin in 36 patients, levetiracetam in 9 patients, clobazam in 8 patients, clonazepam in4 patients, phenobarbital in 2 patients, lamotrigine in 2 patients, and topiramate in 2 patients. The mean number of drugs used during ACTH treatment was 2.06.
Twenty-five (50%) patients used the two most common drugs-both vigabatrin and valproic acid at the same time. During the whole period of ACTH therapy, the children were hospitalized. Table 4. Antiepileptic drugs used before the onset of ACTH.

Patients without Hypsarrhythmia
Study Group

Results
Complete cessation of spasms was achieved in 42 cases out of 50 children (84%) withcombined antiepileptic therapy (ACTH and AEDs). Among children who responded to ACTH, 20 were with hypsarrhythmia. EEG improvement was seen in 41 (82%) patients. Improvement of EEG recording in patiets with hypsarrhytmia meant its withdrawal and, in other cases, reduction of paroxysmal changes and their amplitude and the dominance of slow theta and delta waves. A substantial reduction (75%) in the frequency of seizures was present in five patients (10%). Two of them were seizure-free after 2 weeks of treatment but had recurrence of seizures during a tapering-off period. In patients with genetic etiology, therapy was effective in 5/8 (62.5%) patients. The treatment failed in one patient with tuberous sclerosis (TSC), one with a mitochondrial disease, and one with microdeletion 2q24.4. Efficacy in patients with structural etiology other than TSC was 17/20 (85%). In this group, the treatment was effective in 10/11 patients (91%) with HIE, 5/7 (72%) with cerebral malformations, one patient with two possible causes of epilepsy, and one patient with postinfectious changes. In patients with unknown etiology, the therapy was effective in 20 patients (91%), including all patients who could be classified as cryptogenic according to the previous ILAE classification. A total of 80.5% of patients previously classified as symptomatic had cessation of spasms. The differences between cryptogenic and symptomatic patients in the chi-square test were not statistically significant (p = 0.345) (using of terms symptomatic and cryptogenic, although not recommended by ILAE, allow for better comparison with previous study). There were also no statistically significant differences between the children with known and unknown etiology (p = 0.427). Gender did not have any influence on cessation of seizures (p = 0.501).

Adverse Events during ACTH Treatment
There was no case of death during ACTH therapy. The most common adverse events during treatment and taper were infections (respiratory tract, urinary system, gastrointestinal), N = 26 (52%). Most of them were mild and did not have an impact on the epilepsy treatment. Antibiotics were used in 19 (38%) patients. In six (12%) patients, stereotyped dyskinesia (usually of the facial region: oral buccal) was observed. These movement disorders stopped gradually within a month of discontinuation of ACTH treatment. Arterial hypertension was present in five (10%) patients. Hypertension had been diagnosed by cardiologist, who made the decision of temporary use of blood pressure-lowering drugs. In all patients, the blood pressure values returned to normal after completion of the steroid therapy.

Follow-Up
Information on the clinical course of 28 patients was obtained during phone follow-up. The study was conducted between June and September 2017. We focused on the presence or absence of seizures (not only IS), drug therapy, and development.
The average follow-up was 37 ± 24.4 months. Table 5 shows the duration of follow-up. In 17 (60.7%) cases, seizures were still present. Among patients without seizures, six children were classified as cryptogenic, and five children (23%) were recognized as symptomatic (three children were diagnosed with mitochondrial encephalopathy, one with neurofibromatosis type 1, and one with microdeletion 2q24.4). There was not any significant difference related to gender (p = 1.0).
Normal or near-normal development was observed in 11 out of 28 children (39%).The clinical outcomes of all 28 cases and the comparison of symptomatic and cryptogenic patients are shown in Table 6.
In this study, we present high short-term efficacy of ACTH used concomitantly with other AEDs in a group of patients with IS who did not achieve cessation of seizures or relapsed when taking other AEDs. Monitored adverse events were common (especially infections) but acceptable, and they did not require ACTH discontinuation. Dyskinesia occurred in six (12%) patients. In follow-up, effects were better in cryptogenic patients.
Adverse events in randomized, controlled trials, in which at least one group used hormonal treatment, and were published after 2010 are listed in Table 7.
Using ACTH in polytherapy is an interesting approach, and some studies suggest that it can be more effective than monotherapy in infantile spasms treatment [18][19][20]. A randomized, multicenter, open-label trial provided evidence that hormonal therapy (tetracosactide, prednisolone) and vigabatrin are more effective than hormonal therapy alone during four weeks of observation [18]. A randomized, open-label study from China also suggested that ACTH in polytherapy with MgSO4 has a better effect than ACTH monotherapy in 24 weeks of follow-up [19]. Prospective study at Severance Children's Hospital also demonstrated the potential of a treatment protocol using polytherapy(vigabatrin and prednisolone) for West syndrome [20]. Other studies also showed that initial polytherapy with ACTH and other AED or topiramate followed by low-dose ACTH are effective and well-tolerated options in IS [21][22][23]; however, it is worth noting that the randomized, open-label, controlled trial denied the effectiveness of combination of hormonal therapy(prednisone) and moderate dose of to piramate [24].
The number of patients with cessation of IS in our study is very similar to the groups using hormonal treatments in combination with other drugs [18][19][20]35]. Relatively good short-term efficacy of the add-on ACTH polytherapy in our study might be associated with cumulative effects of different drugs suggested by the previously mentioned studies, especially vigabatrin, which was taken by most patients [18][19][20]35]. Valproate was also commonly used in our study and might have had an important role in the reached outcome, but there are limited data about its efficacy in IS [14,36]. In addition, clobazam used in our study might be a potentially effective adjunctive drug for West syndrome [37]. It is worth mentioning that similar results were reached in some old studies in the case of highdose ACTH monotherapy and combination of low-dose ACTH and pyridoxine [23,38]. According to some authors, increasing the dose of ACTH does not necessarily result in better seizures control [39].
However, on the other hand, we would not favor polytherapy because of the risk of serious adverse effects (e.g., potential hepatotoxicity in infancy). There is also no proven evidence of effectiveness of polyteraphy in infantile spasms. Based on observation, we would like to stress that it is worthwhile to start ACTH even after refractory spasms and relapses and even after a long delay.
Monitored adverse events reported in this study were relatively mild, and they did not require discontinuation of treatment (Table 8).  a If spasm still occurred on day 7, the dose was changed; it reached 15 mg once and was kept at this level for the next 14 days, four times each day. Whether the spasms hadstopped or not, after these 14 days, the dose of prednisone was decreased every week until completing a 49-day or 56-day course (e.g., 40 mg one time a day for a week or 30 mg one time a day for a week, 20 mg one time a day for a week, 10 mg one time a day for a week, 5 mg one time a day for a week, and finally, 5 mg every two days for a week). b Additionally, the combination therapy group used TPM at an initial dose of 1 mg/kg/day twice a day, then gradually increased it to 3 mg/kg/day on day 7 and 5 mg/kg /day on day 14. c No further medications were needed at the end of the therapeutic protocol since symptoms, such as convulsions and spasm, were fully controlled in 18.7% of patients. Apart from ACTH, another medicine was given to 5.37% of patients to fully control the convulsions. Using another drug helped 25% of patients with controlling, to some extent, theoccurring convulsions and spasms. The mentioned symptoms were resistant to treatment in 3.9% of patients despite concomitant treatment with ACTH and several other drugs. There was no significant change in any of the four levels defined for controlling spasms and convulsions. d The method of ACTH administration: 1st week, one time a day; 2nd week, once every second day; 3rd week, two times a week; 4th week, one weekly; and 5th to 8th week, once every two weeks. Increase daily dose if the patient was not responsive after two weeks; 14 days of therapy, and then, the dose was tapered off over the remaining two weeks. e A dose of 8 mg/kg/day of prednisolone (max: 60 mg) was administrated to patients who belonged to the corticosteroid group. It was used for three weeks, divided into three doses. Later, the dose was decreased in responders. Two weeks later, non-responders were given 2-3 U/kg/day of ACTH (max: 100 U) for five days. If the patients did not show signs of improvement within two weeks of treatment, they were planned to receive intramuscular biologic ACTH immediately after the failure of corticosteroid therapy. The patients received five daily doses of 2-3 IU/kg of ACTH (max: 100 IU), and the treatment was then changed to the oral corticosteroid.
In the second group (25 patients), patients first received five daily doses of 2-3 IU/kg of ACTH.    Dyskinesia is a newly reported side effect of hormonal treatment. In this study, six (12%) patients had a dyskinesia similar to that previously reported [40,41]. Many patients in this study took vigabatrin, and movement disorders might be also potentially associated with vigabatrin [42,43]. In ICISS, the group treated with the combination of hormonal therapy and vigabatrin had higher rate of movement disorders than the group taking hormonal therapy alone [18]. In ten investigated patients enrolled in ICISS, eight of them had movement disorders after initiation of vigabatrin, but they resolved after vigabatrin withdrawal only in two patients. Seven of them also took hormonal treatment during the onset of the movement disorders [42]. Due to the ICISS report, mentioned movement disorders were not linked to vigabatrin, and they were likely caused by an underlying neurological disease [18].
In our study, the number of patients with a developmental delay and seizures in a long-term perspective was significantly lower in cryptogenic patients, which is consistent with previous studies [8][9][10][11][12]31]. In total, 39% of them were seizure free, and among them, all were cryptogenic patients (100%), and 22.7% were symptomatic patients. This result resembles long-term outcome in other studies; however, some treatment options used in the studies presented in the Table 6 resulted in better long-term outcome, which can be caused by differences in methodology and time of observation [11,19,[25][26][27]29,30,33,34,44]. Cryptogenic patients seem to have slightly better and symptomatic slightly worse seizure outcome than patients from UKISS and the observational multicenter Belgian study by Mert et al. [10][11][12]. Development was normal or near normal in 39% of patients-83% cryptogenic and 27% symptomatic patients. According to the meta-analysis and systematic review published in 2015, 54.3% of cryptogenic and 12.5% of symptomatic patients with IS had normal or near-normal developmental outcome [9]. In the context of these data, the developmental outcome in our study seems to be good. On the other hand, there is a possibility that assessment of development was stricter in other studies. Shorter time to treatment results in better developmental outcome, especially in cryptogenic patients [8,9,31,45]. Good developmental outcome among cryptogenic patients in our study might be also associated with short lead time to treatment in this group of patients.
It should be mentioned that according to the latest meta-analysis, high-dose prednisolone is not inferior to adrenocorticotrophic hormone [46,47]; indeed, ACTH is less cost-effective, defined as probability of clinical spasm resolution 14 days after treatment initiation, than oral prednisolone (in the United States) [48].
Our study has some weakness and limitations. The first is a partially retrospective character. Others are an inaccessibility of almost half of patients for follow-up and differences in time of follow-up for particular patients. The follow-up method, phone interview, might be less accurate than direct clinical assessment. The study did not assess some ACTH adverse events. This study is not a clinical trial but only a description of experience from our clinic. There was not a control group, so it does not allow for making an accurate comparison for different used drugs or approaches in infantile spasms treatment; it only assesses effectiveness of the add-on ACTH therapy in this group of patients.

Conclusions
ACTH used concomitantly with other AED is a highly effective treatment with acceptable side effects. In accordance with other studies, the long-term outcome was better throughout cryptogenic patients. Randomized, controlled, clinical trials with long-term follow-up are needed to compare effectiveness of ACTH in polytherapy and monotherapy. Dyskinesias as a potential side effect observed in our study group should be investigated in future studies.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data analyzed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest:
The authors declare no conflict of interest.