No Benefit to Afatinib in HER2-Positive Breast Cancer With Brain Metastases

November 18, 2015

The use of afatinib failed to show a benefit in HER2-positive breast cancer patients with progressive brain metastases.

The use of afatinib or afatinib plus vinorelbine during or after treatment with trastuzumab, lapatinib, or both failed to show a patient benefit in women with HER2-positive breast cancer with progressive brain metastases compared with investigator’s choice of treatment, according to the results of a phase II study published in Lancet Oncology.

Afatinib, an oral inhibitor of the ErbB family of proteins, did benefit about one-third of patients assigned to the treatment, but had no better activity than investigator’s choice of therapy, which consisted mostly of trastuzumab or lapatinib plus chemotherapy.

“No objective responses were achieved in patients treated with afatinib alone, but around one-third of patients, including those treated with afatinib alone, did not have disease progression during the first 12 weeks,” wrote researcher Javier Cortés, MD, of Vall d’Hebron Institute of Oncology in Barcelona, and colleagues. “No further development of afatinib for HER2-positive breast cancer is currently planned.”

The multicenter phase II trial included women age 18 or older with confirmed HER2-positive breast cancer and central nervous system (CNS) recurrence or progression that occurred either during or after treatment with trastuzumab, lapatinib, or both. Patients were randomly assigned to treatment with afatinib alone (n = 40), afatinib plus vinorelbine (n = 38), or investigator’s choice (n = 43).

The primary endpoint was patient benefit at 12 weeks, defined as absence of CNS or extra CNS disease progression, no tumor-related worsening of neurologic signs or symptoms, and no increase in corticosteroid use.

Patient benefit at 12 weeks was achieved in 12 patients (30%) assigned afatinib alone (difference vs investigator’s choice: -11.9% [95% confidence interval (CI), -32.9–9.7]; P = .37), 13 patients (34.2%) assigned afatinib plus vinorelbine (difference vs investigator’s choice: -7.6% [95% CI, -28.9–14.2]; P = .63), and 18 patients (41.9%) assigned investigator’s choice.

“The study was not powered to compare the treatment groups, but the results suggest that afatinib monotherapy or afatinib plus vinorelbine did not provide better outcomes than the investigator’s choice of treatment,” the researchers wrote. “No objective responses were documented in the afatinib alone group.”

The researchers found that adverse events were frequent and afatinib-containing regimens seemed to be less well tolerated. Specifically, the most common grade 3/4 adverse event was diarrhea. Diarrhea occurred in 18% of patients assigned afatinib and 24% of patients assigned afatinib plus vinorelbine compared with 5% of patients on investigator’s choice. In addition, 43% of patients assigned afatinib alone required dose reduction compared with 7% of patients given investigator’s choice.

“Overall, no unexpected adverse events were noted and the adverse events that did occur were generally manageable; however, the investigator’s choice regimens seemed to be better tolerated, especially when compared with afatinib plus vinorelbine,” the researchers wrote.

In an accompanying editorial, Nancy U. Lin, MD, of Dana-Farber Cancer Institute in Boston, wrote that this trial provided important lessons about drug development and patient care.

“Designing randomized trials in this setting has been challenged by the absence of an obvious standard-of-care control group, because of the shortage of approved regimens for this indication. Single-group studies have often used the results of the lapatinib trials as a reference point; however, increasing exposure to lapatinib in the clinical setting has made results of contemporary uncontrolled studies difficult to interpret,” Lin wrote. “Cortés and colleagues’ study is the first to take a pragmatic approach by allowing investigators to choose the active control regimen, and thus provides the most robust benchmark we now have available against which to assess novel agents, particularly in the post-radiation setting.”