Early economic modelling of a new pharmacotherapeutic treatment pathway for children with monogenic obesity

Background Early onset obesity in children, almost always accompanied by significant health complications, may be driven by rare genetic variants that influence appetite, metabolism, and nutrient absorption. Traditional treatment approaches are usually insufficient for those with monogenic obesity of this type. Glucagon-like peptide-1 (GLP-1) receptor agonists, such as semaglutide, and related drugs such as melanocortin 4 receptor agonists, have emerged as promising first-line treatments for severe obesity. There is no established protocol or pathway in England for identifying children with monogenic obesity who could benefit from these and similar treatments Methods We undertook early economic modelling to examine the cost-effectiveness, from a health service perspective, of implementing a new pharmacotherapeutic care pathway for the identification and treatment of monogenic obesity in children. We modelled a hypothetical population of children with hyperphagia and body mass index (BMI) three standard deviations above mean values for age and sex. We evaluated the clinical decision to initiate the pathway using a decision tree model with patient quality-adjusted life years (QALYs) and NHS healthcare costs 12 months from an initial clinic visit as outcomes, and calculated incremental cost effectiveness ratios and a cost-effectiveness acceptability curve. Results Both costs and QALYs were higher under further investigation (GBP3,247 and 0.47 QALYs) compared to no further investigation (GBP1,589 and 0.24 QALYs). The incremental cost-effectiveness ratio in the base case was GBP7,133 per QALY. Further examination of these children was therefore likely to be cost effective in this model. Conclusion A decision-tree model suggested that further investigation of severely obese children potentially eligible for treatment with semaglutide is likely to be cost-effective for the NHS. However, this result is associated with uncertainty arising from a lack of evidence for many key model parameters.