Prednisolone Versus Vigabatrin in the First-line Treatment of Infantile Spasms

Overview

Infantile Spasms, is an rare age-specific epilepsy of early infancy. A 2012 American Academy Neurology/ Child Neurology Society practice parameter update on the medical treatment of infantile spasms concluded: adrenocorticotrophic hormone or vigabatrin may be offered for short-term treatment of infantile spasms. There was insufficient evidence to recommend the use of prednisolone, dexamethasone, and methylprednisolone. The cost of ACTH and the side effects of vigabatrin have led to the consideration of alternative medications to treat infantile spasms. The United Kingdom Infantile Spasms Study (UKISS) in 2004, comparing the efficacy of intramuscular synthetic ACTH to high dose oral prednisolone, showed a response rate of 74% for ACTH and 70% for prednisolone. Since the UKISS paper was published, many institutions in the United States and Australia have used oral prednisolone instead of ACTH, partly due to the exorbitant cost of intramuscular ACTH but also its ease of use and better adverse event profile compared to ACTH. Prednisolone and vigabatrin are both oral medications, which can be initiated promptly upon diagnosis of infantile spasms, expediting treatment and shortening treatment lag time. Because the UKISS trial is the only Class 3 study providing evidence for oral prednisolone in the first-line treatment of infantile spasms, further prospective studies are needed.

Background and Rationale for the study: Infantile spasms (IS) is a rare, catastrophic age-specific epilepsy syndrome with onset within the first 12 months of life. The disorder is characterized clinically by epileptic spasms, often accompanied by developmental regression and a characteristic interictal electroencephalography (EEG) pattern called hypsarrhythmia. When all these three features are present, the term, "West syndrome" is commonly used. The catastrophic nature of the disorder is due to the frequent sequelae of severe global neurodevelopmental delay and medically refractory epilepsy. The initial age of onset in 90% of cases occur before 12 months of life, with peak onset at 6 months. The incidence is 2 to 3 per 10,000 live births with a lifetime prevalence of 1.5 to 2 per 10,000 children. It is slightly more common in males, and a family history exists in 3%-6% of cases.1,2,3 The etiology of IS has conventionally been divided into symptomatic or cryptogenic, depending on whether the underlying cause is known. Conditions causing symptomatic IS are diverse and can include focal or multifocal brain injuries, chromosomal abnormalities and genetic mutations and inborn errors of metabolism. In about 20% of cases, the identifiable cause remains unknown and these are defined as cryptogenic. The spontaneous remission rate of IS in limited natural history studies is 30%.4,5 Although the clinical spasms and typical EEG pattern disappear by 3 to 4 years of age, up to 60% of children with IS will go on to develop other types of refractory seizures, including Lennox Gastaut syndrome. The poor developmental outcome in IS is primarily due to the underlying etiology, however, delay in diagnosis and treatment6 and the use of ineffective therapies are also contributory factors. The hypsarrhythmia EEG pattern is a pattern of epileptic encephalopathy and there is evidence that longer duration of epileptic encephalopathy contributes to developmental delay and to the development of autism, especially in infants with Down syndrome and tuberous sclerosis complex (TSC).7,8 Currently, there is still considerable variation in the management of IS, as evidenced by the US Consensus Report and a recent survey done on the current evaluation and treatment of IS among members of the Child Neurology Society (CNS). Most neurologists use adrenocorticotrophic hormone (ACTH) as their preferred, first-line treatment of IS, not caused by TSC and Vigabatrin (VGB) as the first-line treatment of IS caused by TSC. 9,10 The 2004 American Academy of Neurology (AAN) and CNS practice parameter on the medical treatment of IS and the 2012 update of this evidenced-based guideline concluded that ACTH or VGB may be useful for the short-term treatment of IS, with ACTH being more effective than VGB excluding cases with TSC. 11,12 The 2012 update also concluded that there is insufficient evidence to determine whether other forms of corticosteroids are as effective as ACTH. There is also insufficient evidence to recommend other agents, or combination therapy in the short-term treatment of IS. Although many conditions can cause IS, several hypotheses regarding the underlying mechanisms have been proposed. The corticotrophin releasing hormone (CRH) hypothesis is one of the most studied. Children with IS, have the common characteristic of excessive release of CRH. Elevated CRH can cause seizures and neuronal death within the hippocampus and amygdala of immature brains in animal studies. Elevated CRH also cause desensitization of CRH receptors leading to decreased ACTH release and resulting in reduced ACTH levels in the cerebrospinal fluid (CSF). ACTH treatment, most likely suppresses the excessive production of CRH via direct action on melanocortin receptors eliciting an antiepileptic effect.13 Prednisolone is a primary metabolite of a synthetic glucocorticoid, prednisone. Corticosteroids can reduce hippocampal excitability in vitro, increases gamma-aminobutyric acid (GABA) and antagonize 5-hydroxytryptophan.14 VGB is an irreversible, enzyme-activated, selective inhibitor of GABA-transaminase. By inhibiting its catabolism, there is increased availability of GABA within the synaptic cleft, increasing its inhibitory effect.15 The mammalian target of rapamycin (mTOR) pathway is a key signaling pathway that is dysregulated in TSC. Animal studies have shown that VGB partially inhibited mTOR pathway activity and glial proliferation in the knock-out mice in vitro, as well as reduced mTOR pathway activation in cultured astrocytes from both knock-out and control mice. This may account for the unique efficacy of VGB in TSC.16 At our institution, VGB is the preferred first-line treatment for all newly diagnosed IS due to its relative ease of use and fewer acute adverse events compared to ACTH. However, concerns regarding retinal toxicity17 require regular monitoring with electroretinograms (ERGs) performed under sedation.18 Also, VGB-related MRI changes are seen in 22-32% of children treated for IS and although these changes are mostly asymptomatic and may resolve even when VGB is continued, there is concern that the changes reflect a medication-related neurotoxic effect.19,20 For patients who fail VGB, we use a 6-week course of synthetic ACTH (Synacthen) given intramuscularly (IM) every other day as second-line agent. However, it is associated with significant adverse events, which include infections from immunosuppression, arterial hypertension, weight gain, severe irritability, gastric irritation, hyperglycemia, electrolyte disturbances, cerebral atrophy, and behavioral changes. The United Kingdom Infantile Spasms Study (UKISS) demonstrated the superiority of hormonal therapy over VGB for cessation of spasms at 14 days in infants without TSC. The hormonal therapy arm included patients allocated high-dose oral prednisolone and intramuscular ACTH and the study showed that prednisolone was as effective as ACTH.21 Because this is the only Class 3 study providing evidence for oral prednisolone in the first-line treatment of IS, further prospective studies are needed. Since 2010, the Hospital for Sick Children (SickKids) started providing the option of either IM ACTH or high-dose oral prednisolone as second-line treatments when patient failed VGB. Our previous retrospective case review of twenty IS patients who had failed a 2-week course of VGB showed an 80% response rate (12/15) to ACTH and only 20% (1/5) to prednisolone. 22. This seemingly low response to prednisolone may be due to the underlying etiology, which made the condition refractory to VGB in the first place. Since the UKISS paper was published, many institutions in the US and Australia have used oral prednisolone instead of ACTH, partly due to the exorbitant cost of intramuscular ACTH (US$70,000 per course for natural ACTH) but also its ease of use and better adverse event profile compared to ACTH.23,24 A retrospective case series of 17 newly diagnosed IS given high-dose oral prednisolone as first-line treatment showed clinical response of 100% in the cryptogenic group, and 64% in the symptomatic (non-TSC) group.24 A retrospective case series of 27 newly diagnosed IS given high dose prednisolone (8mg/kg/day) with a maximum daily dose of 60 mg showed a 63% response rate (17/27) to prednisolone within 2 weeks. 25 The selection of the starting dose and 4 week, duration of treatment was based on the previous studies, which used a dose range of 4-8 mg/kg/day and 8mg/kg/day respectively with 4 - 6 weeks duration of therapy. 21.23.24.25. We elected to use the higher end of the dose range previously reported to ensure maximum efficacy, with the shorter length of 4 weeks of treatment to reduce the risk of medication side effects. The relapse rate for infantile spasms in retrospective studies has ranged between 12-40%, with time to relapse between 2-25 months. 24,25. We do not envisage using long-term treatment or recurrent intermittent treatment with prednisolone. Prednisolone and VGB are both oral medications, which can be initiated promptly upon diagnosis of IS, expediting treatment and shortening treatment lag time. We propose that oral prednisolone may be more effective than VGB when used as a first-line treatment in newly diagnosed non-TSC IS patients. and Objective of the study: Aim To compare the efficacy of prednisolone and VGB as 1st line short-term treatment of IS as measured by clinical and EEG response at 2 weeks. Hypothesis: More patients allocated to prednisolone therapy will have cessation of spasms and resolution of hypsarrhythmia at 2 weeks. METHODS: Study design: Single center, prospective, observational, open trial using high-dose oral prednisolone as first-line treatment for newly diagnosed IS (non-Tuberous Sclerosis), with follow up at 2 weeks, the end of treatment and then 5 months later. All patients will undergo thorough history, physical and neurological examination, metabolic, genetic, EEG and neuroimaging studies as per the Hospital for Sick Children (HSC) Infantile Spasms Guidelines. 25 These patients will be compared to our historical controls, composed of our cohort of non-TSC IS patients from January 2010- September 2013 who received VGB as first-line treatment, met the inclusion criteria and were followed up for at least 6 months from diagnosis and treatment initiation. The patients and their families are under no obligation to be involved in the study and will continue to receive care from their primary neurologists should they choose not to participate in the study. Study Duration: The length of the study for each patient will be approximately 6 months. The total duration of the study is estimated at 3 years. Sample size: A sample population of 70 patients in the retrospective VGB cohort and 35 patients in the prospective Prednisolone cohort with a 2:1 match are required to demonstrate a statistically significant difference in efficacy of 25% between Prednisolone and Vigabatrin. Patient Population: Children between the ages of 2-24 months with a clinical spasms and hypsarrhythmia or modified hypsarrhythmia on initial EEG referred to the outpatient Neurology clinic at Sickkids. Control Population: Historical controls will be composed of our cohort of non-TSC IS patients from 2010-2013 who received VGB as first-line treatment and fulfill the inclusion criteria for the study.