S4-8: First results of AVEREL, a randomized phase III trial to evaluate bevacizumab (BEV) in combination with trastuzumab (H) + docetaxel (DOC) as first-line therapy for HER2−positive locally recurrent/metastatic breast cancer (LR/mBC)

Abstract

Abstract Background: H + taxane is an established and effective first-line treatment for HER2−positive LR/mBC. Preclinical data provide the rationale for combining BEV and H in HER2−positive LR/mBC but clinical data in this setting are limited to single-arm phase II studies. AVEREL is the first randomized trial of BEV in HER2−positive LR/mBC. Methods: Eligible patients had measurable or evaluable HER2−positive LR/mBC (centrally confirmed IHC 3+ or FISH/CISH +), ECOG performance status 0/1, and had received no prior chemotherapy for advanced disease. Prior adjuvant H/taxanes were permitted unless disease had recurred <6/<12 months after the last dose, respectively. Patients with CNS metastases were excluded. Patients were stratified according to adjuvant H, prior (neo)adjuvant taxane (further stratified by time to relapse since last dose of [neo] adjuvant chemotherapy/no chemotherapy), hormone receptor status, and measurable disease, and were randomized to receive either H (8 mg/kg → 6 mg/kg q3w) + DOC (100 mg/m2 q3w) or H + DOC + BEV (15 mg/kg q3w). H and BEV were given until disease progression; DOC was given for a planned minimum of 6 cycles unless progression or unacceptable toxicity mandated earlier discontinuation. The primary endpoint was progression-free survival (PFS). Additional endpoints included overall survival (OS), overall response rate (ORR; assessed by RECIST 1.0), duration of response, time to treatment failure, safety (NCI CTCAE v3.0 adverse events [AEs] and AEs of special interest for BEV), quality of life (FACT-B), and translational research. The primary analysis of PFS was prespecified after 310 investigator-assessed PFS events. The statistical design provided 90% power to detect a PFS hazard ratio (HR) of 0.69 (median PFS 11→16 months) with α=0.05. Results: Between Sep 2006 and Feb 2010, 424 patients were enrolled from 60 centers; 421 received treatment. Baseline characteristics were generally well balanced in the H + DOC and H + DOC + BEV arms: median age 55 and 53 years, respectively; visceral metastases 71% and 77%; prior adjuvant H 12% and 13%; disease-free interval <12 months 43% in both arms; measurable disease 85% in both arms. Median follow-up was 26 months in both arms. Efficacy results are summarized in the table. No new safety signals were observed. The following grade ≥3 AEs were more common in the BEV-containing arm than the H + DOC arm: congestive heart failure (5.1% vs 2.9%, respectively); febrile neutropenia (11.6% vs 8.7%); hypertension (11.6% vs 0.5%). Grade 5 AEs occurred in 1.4% of the H + DOC + BEV arm vs 1.9% of the H + DOC arm. Conclusions: The addition of BEV to H + DOC improved PFS without reaching statistical significance according to investigator assessment (unstratified HR 0.82; 95% CI 0.65−1.02). The improvement as assessed by Independent Review Committee was statistically significant. Evaluation of biomarkers is ongoing to try to identify those patients who may benefit from first-line BEV-containing therapy for HER2−positive LR/mBC. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr S4-8.