Background: For hormone receptor-positive (HR+) advanced breast cancer (ABC), guidelines recommend endocrine therapy as the preferred option even in the presence of visceral disease. If there is evidence of endocrine resistance or rapidly progressive disease requiring a fast response then chemotherapy is recommended. The randomized controlled trial (RCT) BOLERO-2 demonstrated that everolimus plus exemestane (EVE+EXE) more than doubled median progression-free survival (PFS) compared with placebo+EXE while still maintaining quality of life in HR+/HER2- ABC patients who recurred or progressed during or after non-steroidal aromatase inhibitors (NSAIs). EVE+EXE offers a viable new line of therapy which can delay treatment with chemotherapy. Therefore, comparative evidence of EVE+EXE versus chemotherapies is relevant. In addition, such data will provide useful information for health technology assessment. Objective: 1) Assess the feasibility of network-meta-analysis (NMA) to compare the efficacy of EVE+EXE with chemotherapies in ABC in terms of PFS. 2) If feasible, to conduct the NMA. Methods: A systematic review of EMBASE, Medline, and Cochrane was performed to identify RCTs concerning the efficacy of everolimus, alternative hormonal therapies, and alternative chemotherapies for ABC in terms of PFS in order to facilitate an indirect comparison of everolimus versus chemotherapy. PFS data from published Kaplan-Meier curves for each treatment and RCT were synthesized and indirectly compared with random effects Bayesian Weibull network-meta-analysis (NMA) models. Analyses were performed assuming treatment effects as expressed with hazard ratios vary over time. Results: Based on a search of 6271 citations 24 RCTs were included. Differences across RCTs were identified in terms of age, post-menopausal status, receptor status, prior exposure to hormonal and chemotherapy, and types of hormonal therapies, visceral metastases, and performance status. Despite the limitations due to these differences, a NMA was performed to provide a broad estimate regarding comparative PFS for everolimus versus chemotherapy to inform current practice. Results of NMA in terms of mean PFS (i.e. area under the curve) up to 20 months are presented in Table 1. Sensitivity analyses will be performed to explore the impact of these differences on the robustness of the results. Table 1. Mean PFS until 20 months as obtained with NMATreatmentMean PFS time up to 20 months95% Credible Interval95% Credible IntervalEpirubicin6.162.5312.17Liposomal doxorubicin6.663.1212.00Paclitaxel7.103.7811.90Pegylated liposomal doxorubicin7.143.0913.56Doxorubicin7.353.9612.09Vinorelbine7.552.9914.28Docetaxel7.924.3712.45Capecitabine9.905.0316.67Nab paclitaxel10.105.4215.75Exemestane+ Everolimus12.216.2116.98 Conclusions: It is feasible to conduct NMA, however, differences in patient characteristics identified among studies are likely to have resulted in over or underestimation of the relative treatment effects. Based on the NMA, EVE+EXE is expected to be at least as efficacious as selected chemotherapies regarding PFS. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P6-10-03.

Background: Activation of the PI3K/mTOR pathway is implicated in resistance to trastuzumab. Accordingly, the BOLERO-3 study evaluated the efficacy of adding everolimus (EVE), an mTOR inhibitor, to vinorelbine and trastuzumab. At the final progression-free survival (PFS) analysis, EVE significantly improved PFS vs PBO (hazard ratio [HR] = 0.78; log-rank P = .0067) but EVE-treated patients had higher rate of grade 3/4 toxicity. To further qualify the benefit:risk of adding EVE to trastuzumab-based therapy, per-protocol, patient-reported, health-related quality-of-life (HRQoL) data were analyzed. Methods: BOLERO-3 is a randomized phase 3, double-blind, placebo-controlled, international multicenter trial. Taxane-pretreated patients (N = 569) with trastuzumab-resistant, HER2+, advanced breast cancer were randomized (1:1) to treatment with EVE or placebo (PBO) plus vinorelbine and trastuzumab. The European Organisation for Research and Treatment of Cancer (EORTC) quality-of-life questionnaire C30 (QLQ-C30) (including the breast cancer-specific BR23 module) was administered at baseline and every 6 weeks thereafter until progression. The QLQ-C30 consists of 30 items combined into 15 subscales, including Global Health Status and functional subscales, where higher scores (range, 0 to 100) indicate better HRQoL. Time to definitive deterioration (TTD) based on a 10% decrease from baseline for GHS and for the physical, emotional, and social function subscales was determined using the Kaplan-Meier method. Treatment arms were compared using a 2-sided log-rank test stratified by prior use of lapatinib. Results: Overall, there was no significant difference in median TDD of HRQoL between treatment arms. The median TTD in global health status score was 8.3 months for EVE (95% confidence interval [CI], 6.9-11.5) vs 7.3 months for PBO (95% CI, 5.6-10.4; P = .8386). The median TTD in the physical, emotional, and social function subscale scores showed no significant difference between arms. For example, median TTD in the physical function subscale score was 12.0 months (95% CI, 8.3-14.1) for EVE vs 12.5 months (95% CI, 8.3-20.9) for PBO (P = .4251), and median TTD in the emotional function subscale score was 15.2 months (95% CI, 9.2-17.3) for EVE vs 12.5 months (95% CI, 9.7-16.4) for PBO (P = .8140). Conclusions: These analyses demonstrate that, despite increased frequency of adverse events observed with the addition of EVE to the standard treatment of vinorelbine and trastuzumab, overall and functional HRQoL scores were not negatively impacted in patients with trastuzumab-resistant, HER2+, advanced breast cancer. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P4-12-18.