In a phase III randomized study, patients with metastatic breast cancer who were treated with the novel agent eribulin mesylate lived on average 2.5 months longer than those treated with existing single agents, representing a statistically significant improvement in overall survival (OS) in an international study presented at the 2010 ASCO Annual Meeting.¹

"Eribulin is unique in that it is the only single agent to show a survival benefit in women with heavily pretreated disease," said principal investigator Christopher Twelves, MD, Professor of Clinical Cancer Pharmacology and Oncology at the University of Leeds in the United Kingdom. "The improvement seen is statistically significant and clinically meaningful for these women whose prognosis is poor."

Eribulin mesylate is a synthetic analog of halichondrin B that binds to a unique site on the microtubule via a novel mode of action. It has potent antiproliferative effects, is active against β-tubulin-mutated cell lines, and induces less neuropathy in preclinical models than paclitaxel, according to Dr. Twelves.

The EMBRACE study (Eisai Metastatic Breast Cancer Study Assessing Physician's Choice versus Eribulin) was a global, randomized, open-label phase III trial involving 762 patients with locally recurrent or metastatic breast cancer who had received a median of four prior chemotherapy regimens; all had received prior anthracycline and taxane treatment and over 70% had also received capecitabine (Xeloda). Most cancers were estrogen-receptor-positive and HER2-negative, and about half had more than two sites of metastases.

Patients were randomized 2:1 to eribulin, 1.4 mg/m2 given intravenously on days 1 and 8 every 3 weeks (n = 508), or treatment according to physician's choice (TPC, n = 254). TPC could consist of any single agent-cytotoxic, endocrine, or biologic-approved for the treatment of cancer. Although best supportive care alone was acceptable, no patient received this, or a biologic agent, as sole treatment.  The most commonly employed TPC was vinorelbine, followed by gemcitabine (Gemzar) and capecitabine.

"Our control arm was different from that of most trials. We allowed physician's choice because there is no single established standard, and we felt it was inappropriate to restrict the options. This was a real-life comparison," Dr. Twelves explained at a press briefing.

Overall Survival Improved by 2.5 Months

Median OS, the primary endpoint, improved from 10.65 months with standard single-agent therapy to 13.12 months with eribulin, representing an overall 19% reduction in mortality risk (P = .04); the 1-year survival rate was 53.9% with eribulin and 43.7% in the TPC arm.

Interestingly, progression-free survival (PFS) by independent review was also numerically superior, but this did not reach statistical significance: 3.7 vs 2.2 months with TPC (HR = 0.87; P = .14); by investigator review, however, the difference was highly significant (HR = 0.76; P = .002). Although the reasons for this discrepancy are not entirely clear, Dr. Twelves pointed out that many patients were censored in the independent review because their metastatic site was not radiologically evaluable.

Objective response rates were modest, but overall responses were significantly more common with eribulin: 12.2% vs 4.7% (P = .002) by independent review and 13.2% vs 7.5% by investigator review (P = .028).

Manageable Toxicity

"These benefits were achieved with a manageable toxicity profile," Dr. Twelves announced. Serious adverse events were observed in about 25% of patients in each arm, and adverse events leading to treatment interruption, dose delays, and interruptions and discontinuations were very similar. Fatal adverse events occurred in 4.0% on eribulin and 7.3% on TPC, and were treatment-related in approximately 1% of each. Grade 3 or 4 neutropenia was observed in 45.2% of patients on eribulin and 21.1% of those receiving TPC, and peripheral neuropathy was seen in 8.2% and 2.0%, respectively. "The 8.2% rate of grade 3 or 4 neuropathy is very encouraging for a microtubule-targeted agent, especially since patients with neuropathy up to grade 2 at baseline were eligible," he commented.

"EMBRACE is the first phase III single-agent study in heavily pretreated metastatic breast cancer to meet its primary endpoint of prolonged overall survival, so these are striking findings," Dr. Twelves concluded. "The 2.5-month improvement in median survival represented a 23% benefit. We see these findings as potentially establishing eribulin as a new treatment option for women with heavily pretreated metastatic breast cancer."

Eric P. Winer, MD, Director of the Breast Cancer Program at Dana-Farber Cancer Institute, Boston, commented at a press briefing that a 2.5-month improvement is "a difference that is sufficient to make one look seriously at this agent." ■

Reference

Twelves C, Loesch D, Blum JL, et al: A phase III study (EMBRACE) of eribulin mesylate versus treatment of physician's choice in patients with locally recurrent or metastatic breast cancer previously treated with an anthracycline and a taxane. 2010 ASCO Annual Meeting. Abstract CRA1004. Presented June 8, 2010.