PND57 Cost-Effectiveness Analysis of a Hypothetical Disease Modifying Therapy for Parkinson's Disease

Abstract

To conduct a case study of the lifetime cost-effectiveness of a hypothetical disease modifying therapy (DMT) plus standard of care (SOC) vs SOC for treatment-naïve Parkinson’s disease (PD) in the UK. A patient-level simulation was developed to account for individual variability in progression of PD, and inter-related changes in Movement Disorder Society Revision of Unified PD Rating Scales (MDS-UPDRS I-IV). Predictive equations were based on analyses of Parkinson's Progression Markers Initiative (PPMI) study and National Institute of Neurological Disorders and Stroke Exploratory Trials in PD Long-Term Study 1. The benefits of starting symptomatic treatments (dopaminergic and advanced therapies) and their long-term limitations were simulated. Inputs for mortality, patient and caregiver utilities, and UK costs (2020) were obtained from publicly available sources. Hypothetical DMT was assumed to slow progression in MDS-UPDRS scores by 50% vs SOC. Costs and health outcomes were discounted at 3.5% per annum. Sensitivity analyses explored the main drivers of results. Patient profiles were generated by jointly sampling correlated characteristics observed in PPMI (mean age=61.2 years, male=65%, PD duration=6 months [range 0 – 48], treatment-naïve). Over a lifetime, patients were projected to incur costs of £243,748—with 22% from treatment costs, and 58% non-medical, including nursing home admission. Comparing DMT vs SOC, quality-adjusted life years (QALYs) were 9.0 vs 7.3 (+1.6), and as costs of DMT were not considered, predicted total costs were lowered by 26%. For a subgroup starting treatment earlier (mean PD duration=3 months [range 0 – 6]), benefits and savings were increased: QALY gain of 1.8 and reduction in costs of £57,251 (-32%). Key drivers of cost-effectiveness: efficacy, utilities, non-medical costs, and baseline characteristics. This case study for a hypothetical DMT illustrates how earlier treatment could be beneficial. The flexible model design facilitates exploratory scenarios and assessment of key sources of uncertainty on the results.