Capecitabine is associated with lower chemotherapy-related expenditures than those associated with vinorelbine in women with metastatic breast cancer.

Abstract

Abstract Abstract #6107 Background: There is growing concern about escalating medical expenditures for patients with cancer. Although chemotherapy acquisition costs have been evaluated, costs of chemotherapy-related complications and administration have been ill defined. This study aimed to quantify chemotherapy-related costs and evaluate their impact on total direct medical expenditures of patients with metastatic breast cancer (mBC) receiving capecitabine or vinorelbine monotherapy.
 Methods: Women with mBC who were treated with capecitabine or vinorelbine monotherapy between 1/1/2003 and12/31/2006 were identified from the Thomson Reuters Marketscan® database. Propensity score methods were used to address selection bias and balance cohorts by matching capecitabine users to vinorelbine users. Chemotherapy-related complications (myelosuppression, infection, constitutional symptoms, and GI-related events) were identified from medical and pharmacy claims. The Cox proportional hazards model was used to assess relative risk of complication events, and general linear models were used to estimate monthly total direct medical expenditures and chemotherapy-related costs.
 Results: 203 capecitabine users were matched 1:1 to vinorelbine users. Mean monthly total direct expenditures were significantly lower for capecitabine users vs vinorelbine users ($7032 vs $9460, respectively, P<.0001). The difference was driven by significantly lower costs of complications (P<.0001) and chemotherapy administration (P<.0001) for capecitabine users. Only 31% of the mean monthly total direct medical expenditures during treatment episodes were related to chemotherapy for capecitabine users (16% complication related, 14% acquisition, 1% administration), vs 51% for vinorelbine users (36% complication related, 11% acquisition, 4% administration). Lower complication costs during capecitabine treatment episodes were from reduced frequency of any complication evaluated (HR=0.37, P<.0001). Capecitabine users were at a significantly lower risk for myelosuppression (HR=0.27, P<.0001), constitutional symptoms (HR=0.67, P=.01), and GI events requiring prescription medications or deemed clinically significant (HR=0.27, P<.0001), vs vinorelbine users. Although the proportion of chemotherapy acquisition/administration costs was similar, actual monthly costs were significantly lower for capecitabine users than for vinorelbine users ($1028 vs $1408, respectively, P<.0001). The difference was driven by a significantly lower cost associated with chemotherapy administration for capecitabine users ($42 vs $410, respectively. P<.0001).
 Conclusion: Patients with mBC receiving capecitabine monotherapy had significantly lower risk for chemotherapy-related complications and lower medical expenditures during treatment than those receiving vinorelbine monotherapy. The total cost difference was driven by lower costs of chemotherapy-related complications and chemotherapy acquisition/administration for capecitabine users. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 6107.