Abstract P1-12-04: Carboplatin, nab-paclitaxel and bevacizumab as first-line treatment for metastatic breast cancer.

Abstract

Abstract Background: Bevacizumab added to weekly paclitaxel resulted in improved progression free survival (PFS) and objective response rates (ORR) compared to weekly paclitaxel alone. Nab-paclitaxel and the platins are active in MBC. We conducted an efficacy and safety study of carboplatin, nab-paclitaxel and bevacizumab. Methods: A phase II open label prospective multi-site study enrolled patients (pts) who had measurable MBC according to RECIST 1.1 criteria and no prior chemotherapy for advanced disease. The primary endpoint was PFS with secondary endpoints of overall survival (OS), ORR, and safety. Pts initially received carboplatin AUC 6 day 1, 22,43, plus weekly nab-paclitaxel 100mg/m2 and bevacizumab 15mg/m2 day 1,22,43 of a 56 day cycle. This was later changed to carboplatin AUC 6 day 1, nab-paclitaxel 100mg/m2 day 1,8,15, and bevacizumab 10mg/m2 day 1,15 of a 28 day cycle. Thirty-two pts were required to detect an increase in median PFS from 6.7 to 10.5 mo with 80% power based on a one-sided p = 0.05. Kaplan–Meier analyses estimated PFS and OS. The log rank test was used for the comparison of survival curves between pts with triple negative MBC (TNBC) and pts with non-TNBC. Results: Thirty-two pts were enrolled between 2/2008 and 11/2011 by 1 academic and 5 community oncology practices. Two pts were ineligible due to non-measurable disease and not included in the response analyses. The median age was 58 years (range 35–81), 22 pts (69%) had an ECOG PS 0, 9 (28%) had a PS 1, 1 (3%) PS 2. Twenty-four (75%) pts had ER+ disease, 7 (22%) had TNBC, 1 (3%) had ER-HER2+ disease not eligible for trastuzumab-based therapy. Metastatic sites were bone (26%), liver (18%), loco-regional (16%), and lung (12%). One pt (3%) had bone and loco-regional disease only, 19 (59%) had visceral dominant disease. The median number of weeks on treatment was 28.9 (range 5–131). The median PFS in all pts was 13.6 months (mo) (95% CI 11.2–21.9), with a median OS of 26.8 mo (95% CI 13.3–41.2). The ORR (2 CR and 18 PR) was 66.7% (CI 47.2–82.7). There also were 6 (20%) unconfirmed PR and 3 (10%) stable disease, resulting in a clinical benefit rate of 96.7% (CI 82.78–99.92). There was no significant difference in PFS (median 13.6 vs 16.1mo, p = 0.37) or OS (median 13.6 vs 26.8mo, p = 0.32) in pts with TNBC versus non-TNBC disease. The most common toxicities of any grade (gr) include neutropenia and thrombocytopenia in 24 pts (75%) each, leukopenia and fatigue in 17 pts (53%) each, anemia in 15 (47%), and neuropathy in 10 (31%). Gr 4 neutropenia was seen in 7 pts (22%) without febrile neutropenia, and gr 4 thrombocytopenia occurred in 6 (19%). There were no pts with gr 4 sensory neuropathy. All pts required chemotherapy dose delays, 15 (47%) had chemotherapy dose reductions. Conclusions: The carboplatin, nab-paclitaxel and bevacizumab combination is highly effective with good tolerance in first line MBC. As the role of anti-angiogenic therapy in first line metastatic breast cancer is being clarified, this would be an attractive regimen to test in the (neo)adjuvant setting and together with novel molecular targeted agents. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P1-12-04.