Cost-effectiveness analysis of atezolizumab in advanced triple-negative breast cancer

Abstract

The IMpassion130 trial demonstrated that adding atezolizumab to nanoparticle albumin-bound (nab)-paclitaxel improved the survival of patients with untreated, advanced, programmed death ligand 1 (PDL1)-positive triple-negative breast cancer (TNBC). In view of the high cost of immunotherapy, it is important to examine its value with respect to both benefits and costs. In this study, the cost-effectiveness of atezolizumab/nab-paclitaxel combination therapy relative to nab-paclitaxel monotherapy was evaluated for the first-line treatment of advanced, PDL1-positive TNBC, from a healthcare system perspective. A three-state partitioned-survival model was developed to compare the clinical and economic outcomes of treatment with atezolizumab/nab-paclitaxel combination therapy with nab-paclitaxel monotherapy in patients with advanced TNBC. Clinical data were obtained from the IMpassion130 trial and extrapolated to 5 years. Health state utilities were retrieved from the literature, while direct costs (in Singapore dollars, S$) were sourced from public healthcare institutions in Singapore. The primary outcomes of the model were life years (LYs), quality-adjusted LYs (QALYs), costs and incremental cost-effectiveness ratios (ICERs). One-way and probabilistic sensitivity analyses and scenario analyses were conducted to explore the impact of specific assumptions and uncertainties. Adding atezolizumab to nab-paclitaxel resulted in an additional 0.361 QALYs (0.636 LYs) at an ICER of S$324,550 per QALY gained. The ICER remained high at S$67,092 per QALY even when atezolizumab was priced zero. One-way sensitivity analysis showed that the ICER was most sensitive to variations in the cost of atezolizumab and the time horizon. Scenario analyses confirmed that the ICERs remained high even under extremely favourable assumptions. Given the exceedingly high ICER, adding atezolizumab to nab-paclitaxel was unlikely to represent good value for money for the treatment of advanced PDL1-positive TNBC. Our findings will be useful in informing funding policy decisions alongside other considerations such as comparative effectiveness, unmet need and budget impact.