The cost of drug repurposing: parallel economic evaluation of mirtazapine for severe breathlessness in the multinational BETTER-B trial
摘要
Breathlessness is a prevalent, distressing symptom in advanced respiratory disease. Proven treatments are lacking. Mirtazapine, an antidepressant, is increasingly prescribed. The placebo-controlled BETTER-B trial did not find significant clinical benefit of mirtazapine for severe breathlessness, and patient experiences were mixed. Given mirtazapine’s low cost and wide availability, it is important to understand effects more broadly.
MethodsWe conducted a parallel cost-effectiveness analysis of mirtazapine versus placebo. BETTER-B recruited 225 adults with chronic obstructive pulmonary disease and/or Interstitial Lung Diseases in seven countries between 2021 and 2023. We calculated quality-adjusted life years (QALYs) using EuroQoL EQ-5D-5 L and national value sets, and combined self-reported healthcare and informal care frequencies with unit costs. Primary trial endpoint was at day 56, with sensitivity analysis at day 180 (trial exit).
FindingsIn primary analysis mirtazapine reduced QALYs (-0.006 (95% confidence interval (CI): -0.012 to -0.001)) and was associated with higher total costs (+€231 (95% CI: -218 to + 680)). For willingness-to-pay thresholds of €20,000-€40,000 per QALY, mirtazapine had a 1%-2% likelihood of being cost-effective. These findings were substantively unaffected by sensitivity analyses to timeframe, perspective, outcome measurement, and modelling strategy.
InterpretationRepurposing mirtazapine to treat severe breathlessness negatively impacted patient outcomes while being associated with higher formal and informal costs, and should be discouraged. Off-label prescribing of repurposed medicines without robust evidence risks unnecessary strain on healthcare systems and families. Economic evaluation within testing of repurposed medicines is important. Parallel cost-effectiveness analyses can deliver high-value information even when the efficacy trial finds no effect.