Reflecting time dependency in hazard rates to analyse the cost-effectiveness of letrozole or anastrozole versus tamoxifen as adjuvant therapies for early breast cancer in hormone receptor-positive (HR+) postmenopausal women: The U.S. perspective

Abstract

596 Background: The BIG 1–98 and ATAC randomized controlled trials demonstrated that, in postmenopausal women with hormone receptor-positive (HR+) early breast cancer, initial adjuvant therapy for 5 years with the aromatase inhibitors (AIs) letrozole (LET) or anastrozole (ANA) is superior to tamoxifen (TAM). The table shows the difference in early effects of the 2 AIs. This study shows the cost-effectiveness of 5 years of initial adjuvant therapy with LET or ANA versus 5 years TAM in a postmenopausal HR+ population. Methods: Clinical trial data from BIG-198 and ATAC populated a Markov model to estimate the incremental cost per quality-adjusted life year (QALY) gained of the AIs compared to TAM. Variable annual hazard ratios for time to recurrence over the 5 year treatment period were applied to a pooled set of variable annual hazard rates for TAM. Probabilities of breast cancer event type and adverse events (AEs) (end. cancer, fractures, stroke, VTEs, and bleeding) were based on published results of the clinical trials and U.S. studies. Costs of breast cancer and AE care, and QALY weights were taken from published studies. Incremental costs and QALYs were estimated over the lifetime of HR+ women aged 60 yrs, discounted at 5% annually. Results: LET increases QALYs by 0.22 per patient compared with TAM while ANA patients gain 0.18 QALYs. Mean incremental cost per QALY gained for the AIs compared to TAM were $37,328 for LET and $40,531 for ANA. Probabilistic sensitivity analysis demonstrated that LET and ANA have around a 70% probability of being cost-effective at a QALY value of $50,000. Conclusions: Adjuvant treatment of postmenopausal HR+ women with LET or ANA for 5 years is a cost-effective use of U.S. health care resources compared with TAM. Based on an indirect comparison of mean values, LET is more cost-effective than ANA, with a higher predicted increment in quality adjusted survival. Early Effects in the ATAC and BIG-198 Trials ATAC (No. of events at 2.5 years) BIG 1–98 (No. of events at 2 years) ANA TAM Prevented (% reduction) LET TAM Prevented (% reduction) Recurrences 196 236 40 (-17%) 117 168 51 (- 30%) Distant recurrences 133 143 10 (-7%) 87 125 38 (-30%) Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Expert Testimony Other Remuneration Novartis Novartis Novartis