PND33 A Cost-Effectiveness Analysis of Fenfluramine for the Treatment of Seizures for Patients with Dravet Syndrome (DS) in the UK Setting

Abstract

DS is a rare, genetic epileptic encephalopathy presenting in early childhood, characterised by frequent, severe and pharmaco-resistant seizures; associated with developmental and cognitive impairment, premature death and poor patient and carer quality of life (QoL). As DS is highly heterogeneous, a novel patient-level simulation model was developed to evaluate the cost-effectiveness of fenfluramine - a newly licensed ‘add-on therapy’ for the treatment of DS, in the UK setting. Patient-level data from the placebo arm of fenfluramine registration studies were used to model the number of convulsive seizures and convulsive seizure-free days per 28-day cycle in a DS population generalisable to the UK. Utility values were mapped from patient and carer QoL data from the trials. Carer utility values were included in line with the NICE reference case. Natural and DS-specific mortality rates were derived from literature and UK healthcare resource use data from a DS-specific pathway study. The primary base case analysis compared standard of care (SoC) with either add-on fenfluramine or NICE-recommended cannabidiol (with clobazam), over a lifetime horizon. A robust indirect treatment comparison provided relative treatment effects for SoC, fenfluramine and cannabidiol. The estimated ICER of fenfluramine at the proposed NHS price compared with cannabidiol (list price) is £31,773/QALY. Deterministic and probabilistic sensitivity analyses demonstrated the base case analysis results are robust to parameter uncertainty. Scenario analyses indicate the most influential assumptions are the inclusion of utility values for carers, and doses assumed for cannabidiol. At plausible maximum doses of cannabidiol, fenfluramine dominates cannabidiol. In fully incremental analyses, fenfluramine had a lower ICER vs SoC than did cannabidiol, i.e. fenfluramine extendedly dominated cannabidiol. This novel patient-level simulation model, which appropriately characterises the population heterogeneity in DS, demonstrates fenfluramine provides a clinically and cost-effective alternative to existing ‘add-on therapy’ options for this rare, life-threatening syndrome.