OBJECTIVES: Soft tissue sarcomas (STSs) are rare cancers with poor outcomes for patients with advanced disease (median overall survival [OS] is 12 – 16 months) for which the standard first-line treatment has changed little in 40 years. Recently, the United States (US) Food and Drug Administration conditionally approved olaratumab in combination with doxorubicin (OlaDox) based on a randomized, phase 2 trial in 133 patients (JGDG) that reported a significant OS benefit over single-agent doxorubicin (Dox). We investigated the cost-effectiveness of OlaDox versus Dox and five other standard-of-care regimens for patients with advanced or metastatic STS, from a US payer perspective.
METHODS: A partitioned survival model comprising three health states (Progression-free, Progressed, and Dead) was developed to estimate costs and outcomes over patients’ lifetimes. Efficacy data were based on the JGDG study and a network meta-analysis. Adverse-event rates and costs were from published sources. Progression-free survival was estimated from Kaplan-Meier curves. OS was estimated using parametric functions and age-specific mortality adjusted for STS, assuming no treatment-effect after trial follow-up. One-way sensitivity analyses (OWSAs), probabilistic sensitivity analyses, and scenario analyses were performed to evaluate the uncertainty in all model parameters. Costs and outcomes were discounted at 3% per annum.
RESULTS: The incremental cost-effectiveness ratio (ICER) estimate for OlaDox versus Dox was $105,408 per life-year (LY) saved (95% credible interval: $62,501-$245,354). Mean costs and LYs for OlaDox increased by $133,653 and 1.27, respectively. In a fully incremental analysis, all other regimens were dominated or extendedly dominated. In OWSAs and scenario analyses, the ICER per LY saved ranged from $78,669 to $190,662.
CONCLUSIONS: Results suggested a substantial improvement in OS with OlaDox (1.27 LYs versus Dox), and an ICER of $105,408 per LY. Analyses were based on a small phase 2 trial; an ongoing phase 3 trial is expected to reduce uncertainty in future estimates.
