Retrospective cohort study evaluating treatment pathways through care and potential adverse events in neonatal seizures

Rationale and background:

Seizures are one of the most common neurologic emergencies in neonates, arising in ~3/1000 term live births and are associated with significant mortality and neuro-developmental disability (Soul et al, 2018). In contrast to seizures in older children, most neonatal seizures result from acute symptomatic etiologies rather than epilepsy. The most common etiologies include neonatal encephalopathy caused by hypoxic-ischemia (HIE), focal ischemia affecting one or more vascular territories (stroke), intracranial hemorrhage, infection, cerebral dysgenesis, and metabolic disturbances.(Soul et al, 2018, Venkatesan et al, 2016).Treatment is largely guided by physician preference and tradition, due to a lack of data from well-designed clinical studies. Globally, phenobarbital (PB) is standard of care as first-line therapy in neonatal seizures. Phenobarbital has shown efficacy (response usually defined as 80%to 100% reduction in seizure burden in sec/h) of between 50% and 80% in several small studies. Several other antiepileptic drugs (AEDs) have shown some efficacy as first-line treatment eg, levetiracetam (LEV)vs. PB as first-line(Rao et al, 2018)or as adjunct treatment, but gold standard evidence of efficacy is lacking for all compounds and all AED based treatments are currently off-label(Soul et al, 2018). 

While the FDA has requested a neonatal study of lacosamide (LCM, Vimpat®), it is only indicated for the treatment of partial-onset seizures (POS) in patients 4 years of age and older and this indication is currently limited in the US to Vimpatoral dosage forms (Film-2 Coated Tablets and Oral Solution). Vimpat injection is only indicated for the treatment of POSin adults (17 years of age and older) in the US.Count data of LCM prescribing in neonates indicates scant usage for prophylaxis only: PEDSnet database n=27, Soul collaboration dataset n <50, a paper by Glass and colleagues indicates no usage (Glass et al, 2017), and an internal survey shows no use in acute seizure and rarely in prophylaxis. Based on compound characteristics (no titration needed, improved safety, and some evidence of efficacy of UCB compound levetiracetam [Keppra®]), UCB instead proposed to initiate a clinical study in neonatal seizures with the newer sister compound brivaracetam (Briviact®),but this study is encountering serious recruitment challenges. Based on these factors, and due to scant literature in the neonatal seizure population, with variations in the care pathway with numerous AEDs utilized at varying doses according to underlying condition (Rennie et al, 2018;Glass et al, 2017; Venkatesan et al, 2016; Dizon et al, 2019),UCB has decided to set up a large epidemiological neonatal seizure study evaluating AED treatment pathways and outcomes. The study will include all neonates that carry a seizure code and have been treated with any AED from time of first prescription. Patients exposed to oral or ivUCB drugs of LCM, BRV, and LEV will also have serious adverse drug reactions reported only if they were attributed by the original treating physician.

Research question and objectives:

Aim: to understand treatment pathway, outcomes, and safety in neonates treated for seizure.

Primary objective: The study will evaluate AED treatment pathway including dose, time on drugs, refractory proportion of patients per line of therapy, seizure recurrence and rate, etiology of the seizure, comorbidity, and concomitant prescribing. 

Secondary Objectives: Based on the characteristics of the three UCB compounds (LCM, LEV, BRV) with known safety risk profiles in adults and older children, the following events will be evaluated to assess similar risk in neonates. Many of these are based on the LCM safety profile which includes a larger number of risks compared to either LEV or BRV.The body systems reflect appropriate clinical risk in neonates and have been checked by Dr.Janet Soul (external key opinion leader [KOL]).

•To estimate the incidence of the selected medical events for 6 System Organ Classes (SOC) namely: i) Cardiac disorders; ii) Skin and subcutaneous tissue disorders; iii)Nervous system disorders; iv)Metabolism and nutrition disorders; v)Administration site conditions and injury; vi) blood dyscrasias.

•Plus2 Standardized MedDRA Queries (SMQs) namely:i) Drug rash -reaction with eosinophilia and systemic symptoms (DRESS) syndrome; ii) Severe cutaneous adverse reactions.

•Additionally, incident events (morbidities) that occur during exposure to any of the UCB drugs will be classed as adverse events and included in the the study report as safety tables.Where it is unclear if an AE has been classified as an ADR by the original treating physician, the record will be pulled and evaluated including all available laboratory data. This ensures that unique events are not missed.If the event is classified as an ADR, regulatory reporting will occur in the usual way via the links on page 3via local safety and PV using the correct form.FDA do not require expedited reporting of historical data and usually only require one final report of such events at study completion (FDA, 2013).