Eribulin Does Not Substantially Increase Pathologic Complete Response in Phase II Trial of Locally Advanced Breast Cancer

Researchers evaluating the neoadjuvant use of eribulin (E) in treating women with locally advanced breast cancer (LABC) have found that it did not perform as well as paclitaxel, which was used as a control in the trial.

“In early breast cancer, especially in the neoadjuvant setting, eribulin is not more active than paclitaxel, as per our study,” says the study’s lead author Jame Abraham, MD, Co-Director of Cleveland Clinic’s Comprehensive Breast Cancer Program, Director of the Breast Medical Oncology Program and Cleveland Clinic’s principal investigator for the NSABP Foundation.

The study determined that the number of patients who achieved pathologic complete response (pCR) for E was below the threshold for additional research.

“When E was substituted for weekly paclitaxel (WP) in this study, it did not improve pCR to the extent of the control drug in HER2-negative LABC,” the study concludes. “Further study of E with this dose schedule in the neoadjuvant setting does not appear to be warranted.”

There is also an economic consideration, says Dr. Abraham. “Paclitaxel is one of the good old drugs, and the study demonstrates that the older drugs are still as effective as the new drugs, which tend to be more expensive. So in a value-based environment, we will consider paclitaxel the preferred agent.”

Both regimens were equally well-tolerated, with no unexpected toxicities, according to the study. Results were published in Breast Cancer Research and Treatment.

Advantages of neoadjuvant chemotherapy

Researchers estimate that more than 234,000 American women will be diagnosed with breast cancer in 2015, and about a third of those will present with stage IIB and III locally advanced disease. Those at greater risk will most likely be treated with neoadjuvant chemotherapy, for a number of reasons:

• It allows the physician to determine tumor response to a specific treatment regimen while increasing the potential for breast-conserving surgery.
• It downstages axillary tumors.
• Prognosis-based pCR can be used as a surrogate endpoint for outcome.
• Perioperative tissue can be analyzed for molecular and genetic correlates of response.

Comparing newer agents that had previously performed well

The study aimed to compare two agents that had performed well in recent trials: paclitaxel and E.

Paclitaxel was approved in 1998 for treatment of non-small cell lung cancer, and paclitaxel albumin-stabilized nanoparticle formulation, also known as nanoparticle paclitaxel, was approved in late 2013 for the treatment of breast cancer after failure of combination therapy for metastatic disease, or within six months of adjuvant chemotherapy.

E is a non-taxane that binds directly with tubulin, disrupting mitotic spindles, and it inhibits microtubule polymerization. Because of its unique interaction with tubulin, study investigators hypothesized that it might have greater efficacy than other microtubule-targeting therapies in current use. It was approved in 2010 for treatment of late-stage breast cancer.

Dr. Abraham says E performed well in vitro. “It was found to be highly active in the metastatic setting also,” he says.

E’s safety profile was also attractive: In a phase I study of 33 patients, the most common grade 3-4 toxicity was neutropenia in 64 percent of patients; all other adverse events occurred in fewer than 5 percent. A phase III study comparing E with the treatment of physician’s choice found overall survival increased to 2.47 months with E.

Considering E’s activity in the advanced disease setting and its acceptable side-effect profile, E appeared to investigators to be a good candidate to test in both earlier metastatic disease and in the neoadjuvant setting.

Creating a cooperative multi-center trial

The NSABP Foundation developed a clinical trial, NSABP FB-9, to compare the two drugs. It was a phase II, multi-center, randomized study of WP or E followed by doxorubicin and cyclophosphamide (AC) as neoadjuvant therapy for women age 18 and older with HER2-negative LABC including stages IIB and IIIA-IIIC.

Patients were assigned randomly in a 1:2 ratio to either WP or E followed by AC, “… not as a direct comparator but rather to determine that the study population was consistent with those of earlier studies reported in the literature,” according to the study.

Patients were randomly assigned to receive either WP (n = 19) for 12 treatments or E (n = 31) every three weeks for four cycles followed in either case by AC every three weeks for four cycles before surgery. Seventeen of 19 patients who took WP and 25 of 30 who took E completed all cycles.

Eligibility requirements included a diagnosis of invasive adenocarcinoma of the breast as determined by biopsy and a palpable mass in the breast or axilla; either hormone-receptor-positive or hormone-receptor-negative tumors; and HER2-negative status. Patients with inflammatory breast cancer were eligible without measurable disease. Patients were required to have an Eastern Cooperative Oncology Group (ECOG) performance status of 0-1.

Summarizing the trial’s results

The trial’s primary aim was to determine the rate of pCR in the breast and nodes of patients with LABC who received upfront E. The rate of pCR in breast and nodes in neoadjuvant studies using similar eligibility criteria has ranged from 16 to 26 percent, with HER2-negative and hormone-receptor-negative patients (triple-negative) having a higher pCR than HER2-negative hormone-receptor-positive patients. The threshold of interest in this trial was a rate of pCR breast and nodes in excess of 20 percent.

With WP, pCR in breast and nodes was achieved in five of 19 patients (26 percent). Treatment with E resulted in a pCR in breast and nodes in five of 30 patients (17 percent). The number of patients with triple-negative disease was small in both arms.

The study found that the lack of any notable differences in clinical responses, MRI responses, and rates of breast-conserving surgery between WP and E regimens was consistent with the pCR findings.