BOLERO-3 Biomarker Study Provides Clues for Response to Everolimus

Exploratory analysis of biomarkers in the BOLERO-3 trial suggests that the addition of everolimus to trastuzumab plus vinorelbine for HER2-positive advanced breast cancer may be most beneficial in patients with low PTEN or high pS6 levels.

Guy Jerusalem, MD, PhD

Exploratory analysis of biomarkers in the BOLERO-3 trial suggests that the addition of everolimus to trastuzumab plus vinorelbine for HER2-positive advanced breast cancer may be most beneficial in patients with low PTEN or high pS6 levels. In addition, no clear benefit of everolimus was observed in patients with normal PTEN or low pS6 levels, and no treatment—biomarker interaction was seen between everolimus and PI3K mutations.

“Our findings are consistent with the observation that mTOR inhibition attenuates PI3K pathway—activated trastuzumab resistance. These results should be considered exploratory and need further confirmation in a larger group of patients,” said Guy Jerusalem, MD, PhD, of the Centre Hospitalier Universitaire Sart Tilman Liege and the University of Liege, Belgium, in announcing the results at ECC 2013. Additional biomarker analysis of tumor samples from BOLERO-3 is being performed.

BOLERO-3 was a Phase III international trial of locally advanced or metastatic HER2-positive breast cancer treated with a prior taxane. All women had tumors that were resistant to trastuzumab. Subjects were randomized 1:1 to treatment with everolimus, vinorelbine, and trastuzumab or vinorelbine plus trastuzumab and placebo and were treated until disease progression. The addition of everolimus was associated with prolonged progression-free survival (PFS) versus placebo. The absolute benefit in PFS for everolimus was 5 weeks: median PFS was 7 months in the everolimus group versus 5.78 months in the placebo group.

Investigators hypothesized that patients resistant to trastuzumab would be more likely to respond to the addition of everolimus to chemotherapy plus trastuzumab.

This study was based on archival tumor samples from 283 of the 569 patients randomized in the trial, which comprised 50% of the intent-to-treat (ITT) population; 80% of the samples were derived from primary tumors. The samples were assessed for PTEN and pS6 levels by immunohistochemistry. PI3-K gene mutations in exons 9 and 20 were assessed by DNA sequencing.

In patients with low PTEN levels (optimal H-score cut-point, <20th percentile), greater benefit was seen with everolimus, with a median PFS absolute gain of about 18 weeks. Median PFS was 41.9 weeks with everolimus versus 23.1 weeks with placebo. Jerusalem called the absolute gain of 18-19 weeks “clinically meaningful.”

Patients with high pS6 level (optimal H-score cut-point >75th percentile) derived greater benefit from everolimus. Median PFS in those with high pS6 levels was 29.4 weeks on everolimus versus 17.1 weeks for placebo. Patients with low pS6 or normal PTEN had no apparent benefit from the addition of everolimus, however.

No significant biomarker—treatment interactions were observed with PI3K mutations, but this was a very small population of only 15 patients, Jerusalem told the audience.

Clifford Hudis, MD, current President of the American Society of Clinical Oncology and chief of the Breast Cancer Service at Memorial Sloan-Kettering Cancer Center in New York City, and a discussant on the trial, said that in his view, the biomarker analysis results from BOLERO-3 are hypothesis-generating at best. “We are splitting up a nearly negative trial, and we have to take these biomarker results with a grain of salt,” he added.

Jerusalem G, Andre F, Chen D. Evaluation of everolimus (EVE) in HER2+ advanced breast cancer (BC) with activated PI3K/mTOR pathway: Exploratory biomarker observations from the BOLERO -3 trial. Presented at: European Cancer Congress 2013; September 27-October 1, 2013; Amsterdam, Netherlands. Abstract LBA16

<<<

View more from the 2013 European Cancer Congress