OBJECTIVE

This study assessed the cost-effectiveness and budget impact of third-generation chemotherapy regimens with prophylactic granulocyte colony-stimulating factor (G-CSF) relative to second-generation regimens without prophylactic G-CSF for patients with high-risk early breast cancer in Japan.

METHODS

We conducted a cost-effectiveness analysis with Markov modeling and calculated incremental cost-effectiveness ratios (ICERs) for the comparison between second-generation regimens without prophylactic G-CSF and third-generation regimens with prophylactic G-CSF. The comparisons consisted of fluorouracil, doxorubicin, and cyclophosphamide, a second-generation regimen, versus docetaxel, doxorubicin, and cyclophosphamide (TAC) with G-CSF, a third-generation regimen; and doxorubicin, cyclophosphamide, and paclitaxel (AC-T) q3wk, a second-generation regimen, versus dose-dense (DD) AC-T q2wk with G-CSF, a third-generation regimen. Patients were stratified by the age at which chemotherapy was started into cohorts aged 35, 45, and 55 years. Outcomes were estimated in terms of life-years (LYs) and quality-adjusted LYs (QALYs). ICER calculations were done from a societal perspective. We also estimated the budget impact, which included the additional public medical expenditures that would cover all subsequent changes after the additional cost of choosing third-generation regimens if G-CSF were approved for use in third-generation regimens for breast cancer. Costs were calculated using prescription drug prices as of 2006.

RESULTS

Estimated ICER values for TAC with prophylactic G-CSF were yen956,471/LY and yen919,443/ QALY for age 35 years, yen1,125,540/LY and yen1,078,967/QALY for age 45 years, and yen1,302,746/LY and yen1,224,896/QALY for age 55 years. Values for DD AC-T q2wk with prophylactic G-CSF were yen291,931/LY and yen311,232/QALY for age 35 years, yen357,354/LY and yen380,148/QALY for age 45 years, and yen377,011/LY and yen399,761/QALY for age 55 years. TAC or DD AC-T q2wk with prophylactic G-CSF would yield cost savings compared with the respective second-generation regimens if the per-dose cost of G-CSF decreased from yen31,355 to yen15,700 (TAC) or to yen24,300 (DD AC-T). The estimated budget impact is yen9.5 to yen11.0 billion per year for the next 5 years.

CONCLUSION

According to a Markov model for patients with high-risk early breast cancer in Japan, third-generation regimens with prophylactic G-CSF will yield improved outcomes at a greater cost, but estimated ICER values are still less than the suggested cost-effectiveness threshold value of yen6 million (US $60,000, assuming US $1 = yen100) for a gain of 1 QALY.