Comparative efficacy of everolimus plus exemestane versus fulvestrant for hormone-receptor-positive advanced breast cancer following progression/recurrence after endocrine therapy: a network meta-analysis

Abstract

Postmenopausal women with advanced breast cancer recurring/progressing on or after initial (adjuvant or first-line) endocrine therapy may be treated multiple times with one of several endocrine or combinatorial targeted treatment options before initiating chemotherapy. In the absence of direct head-to-head comparisons of these treatment options, an indirect comparison can inform treatment choice. This network meta-analysis compared the efficacy of everolimus plus exemestane with that of fulvestrant 250 and 500 mg in the advanced breast cancer setting following adjuvant or first-line endocrine therapy. The reported hazard ratios (HRs) for progression-free survival (PFS) or time to progression from six studies that formed a network to compare everolimus plus exemestane (BOLERO-2 trial) with fulvestrant were analyzed by means of a Bayesian network meta-analysis. In the primary comparison (PFS analysis based on the local review of disease progression from BOLERO-2 with the data from the other studies), everolimus plus exemestane appeared to be more efficacious than both fulvestrant 250 mg (HR = 0.47; 95 % credible interval [CrI] 0.38–0.58) and 500 mg (HR = 0.59; 95 % CrI 0.45–0.77). Similar results were obtained in an alternate comparison based on central review of disease progression from BOLERO-2 with the data from the other studies (HR = 0.40; 95 % CrI 0.31–0.51 and HR = 0.50; 95 % CrI 0.37–0.67, respectively), and in a subgroup analysis of patients who had received prior aromatase inhibitor therapy (HR = 0.47; 95 % CrI 0.38–0.58 and HR = 0.55; 95 % CrI 0.40–0.76, respectively). These results suggest that everolimus plus exemestane may be more efficacious than fulvestrant in patients with advanced breast cancer who progress on or after adjuvant or first-line therapy with a nonsteroidal aromatase inhibitor.