Randomized phase II trial of nanoparticle albumin-bound paclitaxel in three dosing schedules with bevacizumab as first-line therapy for HER2-negative metastatic breast cancer (MBC)

Abstract

1006 Background: Nanoparticle albumin-bound paclitaxel (nab-P) 260 mg/m2 is superior to paclitaxel 175 mg/m2 (P) every 3 weeks (Gradishar et al., J Clin Oncol. 2005). Weekly uninterrupted P is more effective than q3wk P in MBC (Seidman et al., J Clin Oncol. 2008). Bevacizumab (bev) nearly doubles response rate and time to progression (TTP) when added to P as 1st line therapy for MBC (Miller et al., N Engl J Med. 2007). Methods: This open-label, phase II study randomized patients (pts) to nab-P at 260 mg/m2 q3wk (arm A) versus 260 mg/m2 q2wk with filgrastim (arm B) versus 130 mg/m2 weekly uninterrupted, all with bev (15 mg/kg q3wk arm A, 10 mg/kg q2wk arms B and C). Pts were required to have measurable, HER-2-negative MBC and no prior chemotherapy for MBC. The primary endpoints were response rate and toxicity. Results: Accrual is complete, with 25% of pts still on study as of December 1, 2008. Of 208 pts randomized, 202 (72 arm A, 54 arm B, 76 arm C) were treated, with balanced demographics and baseline characteristics. The median age was 56 (range 29–85). 89% had visceral disease and 61% had prior neo-adjuvant or adjuvant chemotherapy. No significant differences in confirmed complete and partial response rates were noted (A: 42%, B: 42%, C: 41%). TTP was longer in arm C (9.2 months) versus both arms B (6.4 months) and A (7.7 months), overall p = 0.028. As per protocol-specified stopping rule, arm B was closed early due to unacceptable safety profile with significantly more grade ≥ 2 fatigue (B:57%, A: 39%, C:39%, p = 0.048) and bone pain (B: 19%, A: 10%, C:4%, p = 0.024). Neurotoxicity grade ≥ 2 was equivalent across all 3 arms (50%); febrile neutropenia occurred in <2% of pts in all arms. Arm C had significantly less arthralgia, myalgia, and nausea compared with arms A and B. Conclusions: Significant antitumor activity was observed in all arms. Weekly nab-P with bev (Arm C) resulted in a significantly longer TTP. Weekly nab-P with bev (Arm C) appears to have the highest therapeutic index, however sensory neuropathy is limiting, suggesting a 3 week on/1 week off schedule could be preferable and should be studied comparatively. [Table: see text]