OBJECTIVES: DMD is characterized by high medical and indirect costs and low quality of life during later stages of the disease under current standard of care. This study examines the potential for new DMD treatments (e.g., eteplirsen, golodirsen) to meet cost-effectiveness standards using a newly developed loss-of-ambulation (LOA) curve and previously published cost and utility estimates.

METHODS: A Markov model based on 4 DMD-specific health states – early ambulatory (EA), late ambulatory (LA), early non-ambulatory (ENA), and late non-ambulatory (LNA) – was constructed. Time-varying transition probabilities from LA to ENA were derived from a newly developed LOA curve. Health state utilities, direct medical costs, and indirect costs were drawn from published literature. Cost-effectiveness of a hypothetical treatment for DMD that extends life was examined by modifying time spent in each of the first three health states.

RESULTS: The hypothetical treatment produced the lowest incremental cost-effectiveness ratio (ICER) when only time spent in ambulatory states increased. A treatment that extends ambulation by 3.5 years produced an additional 2.02 discounted life years (LYs) and 1.63 QALYs. When treatment also extended ENA by 2 years, the ICER increased substantially due to high costs and low quality of life associated with ENA, shown by a further QALY gain of only 0.31. Relative to an identically priced treatment that extends time only in EA/LA, a treatment that also extends ENA by 2 years will have an ICER of over $250,000.

CONCLUSIONS: Within the current framework, new DMD treatments with life-extending benefits perform poorly on ICER, even without incremental drug cost, due to the existing high cost of care and low quality of life during the later stages of DMD. Widespread application of ICER thresholds may limit access to beneficial treatments, illustrating the need for new methods of assessing the cost effectiveness of treatments for pediatric degenerative diseases.