Meeting News CoveragePerspective

ESR1 mutations linked to poorer OS in ER-positive breast cancer

SAN ANTONIO — ESR1 mutations occurred frequently in the cell-free DNA of women with ER-positive breast cancer and appeared associated with shorter OS, according to an analysis of data from the BOLERO-2 trial presented at the San Antonio Breast Cancer Symposium.

Further, women who harbored the ESR1 D538G mutation benefited from everolimus (Afinitor, Novartis), whereas women with the ESR1 Y537S mutation did not.

Sarat Chandarlapaty, MD, PhD, a breast medical oncologist at Memorial Sloan Kettering Cancer Center, and colleagues have identified mutations in the estrogen receptor in women with advanced breast cancer, especially those who have been exposed to aromatase inhibitors.

Sarat Chandarlapaty, MD, PhD

Sarat Chandarlapaty

“These mutations make the receptor no longer need estrogen — the breast cancer seems to become estrogen independent and cells can grow despite the absence of estrogen,” Chandarlapaty said during a press conference. “That led us to two major questions as we were thinking about this for the clinic: How often do we see these constituently actively estrogen receptor mutations in patients, and how do patients fare who have these mutations present?”

To explore these questions, Chandarlapaty and colleagues evaluated data from 541 women enrolled on the BOLERO-2 trial (74.7% of study population) because women in that trial had ER-positive metastatic disease, were treated with an aromatase inhibitor and had available plasma samples.

Researchers evaluated patients’ plasma, which had been collected at the time of study entry, using QIAamp Circulating DNA kit (Qiagen) or QIAsymphony DSP Virus/Pathogen kit (Qiagen).

Overall, 156 (28.8%) of the women harbored an ESR1 mutation in D538G (21.1%) and/or Y537S (13.3%). Thirty women harbored both types of ESR1 mutations.

Chandarlapaty and colleagues then compared these findings with the presence of ESR1 mutations in 302 archived tumor biopsies — 244 from primary tumors and 57 from metastases — from the cohort. Results showed only four (1.3%) samples had detectable ESR1 D538G and one (0.3%) had ESR1 Y5375.

There are three reasons why ESR1 mutations may be more common in plasma than tumor biopsies, Chandarlapaty said.

“The archival tumor is often the primary breast cancer, and we’ve already observed that with metastases, there is more frequent identification of mutations,” he said. “By virtue of the study, we’re only looking at the time they had metastatic disease when we’re looking at the plasma.”

Further, primary tumors often were collected prior to aromatase inhibitor exposure.

“So, there is a selection over time for these active mutants when you give them a drug,” Chandarlapaty said. “After they’ve been exposed to the drug, there is a higher incidence of the mutation.

“Third, when we evaluate plasma, we are looking at the summation of metastatic breast cancer, not just at one particular lesion that we biopsied, but perhaps many different lesions,” he added.

When researchers evaluated outcomes, median survival appeared shorter among women who harbored the ESR1 D538G mutation (26 months; HR = 1.25; 95% CI, 1.02-1.54), Y537S mutation (20 months; HR = 2.31; 95% CI, 1.34-3.97) or both (15.2 months; HR = 1.77; 95% CI, 1.31-2.39) compared with patients with wild-type ESR1 (32.1 months).

Researchers then assessed how these mutations affected outcomes with the addition of everolimus to exemestane on the BOLERO-2 trial.

Everolimus appeared to prolong median PFS among wild-type patients (8.5 months vs. 3.9 months; HR = 0.4; 95% CI, 0.31-0.51) and patients with the ESR1 D538G mutation (5.8 months vs. 2.7 months; HR = 0.34; 95% CI, 0.2-0.57). However, patients with the ESR1 Y537S mutation did not derive benefit from everolimus (4.2 months vs. 4.1 months; HR = 0.98; 95% CI, 0.49-1.94).

“This was a small group of patients in the Y537S group,” Chandarlapaty said. “We wouldn’t today say, ‘Don’t go out and give [everolimus],’ based on this result. But it does tell us the biology, and what we need to do moving forward in the clinic is look at these mutations, not just as a whole, but individually, and ask how they impact therapies that we are developing.” – by Alexandra Todak

Reference:

Chandarlapaty S, et al. Abstract S2-07. Presented at: San Antonio Breast Cancer Symposium; Dec. 8-12, 2015; San Antonio.

Disclosure: Chandarlapaty reports grant support from Novartis and consulting fees from Chugai, Foresite and Oncothyreon.

SAN ANTONIO — ESR1 mutations occurred frequently in the cell-free DNA of women with ER-positive breast cancer and appeared associated with shorter OS, according to an analysis of data from the BOLERO-2 trial presented at the San Antonio Breast Cancer Symposium.

Further, women who harbored the ESR1 D538G mutation benefited from everolimus (Afinitor, Novartis), whereas women with the ESR1 Y537S mutation did not.

Sarat Chandarlapaty, MD, PhD, a breast medical oncologist at Memorial Sloan Kettering Cancer Center, and colleagues have identified mutations in the estrogen receptor in women with advanced breast cancer, especially those who have been exposed to aromatase inhibitors.

Sarat Chandarlapaty, MD, PhD

Sarat Chandarlapaty

“These mutations make the receptor no longer need estrogen — the breast cancer seems to become estrogen independent and cells can grow despite the absence of estrogen,” Chandarlapaty said during a press conference. “That led us to two major questions as we were thinking about this for the clinic: How often do we see these constituently actively estrogen receptor mutations in patients, and how do patients fare who have these mutations present?”

To explore these questions, Chandarlapaty and colleagues evaluated data from 541 women enrolled on the BOLERO-2 trial (74.7% of study population) because women in that trial had ER-positive metastatic disease, were treated with an aromatase inhibitor and had available plasma samples.

Researchers evaluated patients’ plasma, which had been collected at the time of study entry, using QIAamp Circulating DNA kit (Qiagen) or QIAsymphony DSP Virus/Pathogen kit (Qiagen).

Overall, 156 (28.8%) of the women harbored an ESR1 mutation in D538G (21.1%) and/or Y537S (13.3%). Thirty women harbored both types of ESR1 mutations.

Chandarlapaty and colleagues then compared these findings with the presence of ESR1 mutations in 302 archived tumor biopsies — 244 from primary tumors and 57 from metastases — from the cohort. Results showed only four (1.3%) samples had detectable ESR1 D538G and one (0.3%) had ESR1 Y5375.

There are three reasons why ESR1 mutations may be more common in plasma than tumor biopsies, Chandarlapaty said.

“The archival tumor is often the primary breast cancer, and we’ve already observed that with metastases, there is more frequent identification of mutations,” he said. “By virtue of the study, we’re only looking at the time they had metastatic disease when we’re looking at the plasma.”

Further, primary tumors often were collected prior to aromatase inhibitor exposure.

“So, there is a selection over time for these active mutants when you give them a drug,” Chandarlapaty said. “After they’ve been exposed to the drug, there is a higher incidence of the mutation.

“Third, when we evaluate plasma, we are looking at the summation of metastatic breast cancer, not just at one particular lesion that we biopsied, but perhaps many different lesions,” he added.

When researchers evaluated outcomes, median survival appeared shorter among women who harbored the ESR1 D538G mutation (26 months; HR = 1.25; 95% CI, 1.02-1.54), Y537S mutation (20 months; HR = 2.31; 95% CI, 1.34-3.97) or both (15.2 months; HR = 1.77; 95% CI, 1.31-2.39) compared with patients with wild-type ESR1 (32.1 months).

Researchers then assessed how these mutations affected outcomes with the addition of everolimus to exemestane on the BOLERO-2 trial.

Everolimus appeared to prolong median PFS among wild-type patients (8.5 months vs. 3.9 months; HR = 0.4; 95% CI, 0.31-0.51) and patients with the ESR1 D538G mutation (5.8 months vs. 2.7 months; HR = 0.34; 95% CI, 0.2-0.57). However, patients with the ESR1 Y537S mutation did not derive benefit from everolimus (4.2 months vs. 4.1 months; HR = 0.98; 95% CI, 0.49-1.94).

“This was a small group of patients in the Y537S group,” Chandarlapaty said. “We wouldn’t today say, ‘Don’t go out and give [everolimus],’ based on this result. But it does tell us the biology, and what we need to do moving forward in the clinic is look at these mutations, not just as a whole, but individually, and ask how they impact therapies that we are developing.” – by Alexandra Todak

Reference:

Chandarlapaty S, et al. Abstract S2-07. Presented at: San Antonio Breast Cancer Symposium; Dec. 8-12, 2015; San Antonio.

Disclosure: Chandarlapaty reports grant support from Novartis and consulting fees from Chugai, Foresite and Oncothyreon.

    Perspective
    Carlos L. Arteaga

    Carlos L. Arteaga

    Based on what Chandarlapaty and colleagues reported, what we know about this mutation from the laboratory is that they allow the receptor to function in the absence of estrogen. The implication is that these mutations mark tumors that do not respond to aromatase inhibitors anymore.

    It is important to know that we do not have much actionability for this information, except to maybe enroll patients in some clinical trial.

    The other thing that is important to emphasize is that this is a not a blood cancer test. This is a test that picks up a mutation, it is not a test that will tell you whether a person has cancer or not.

    • Carlos L. Arteaga, MD
    • Vanderbilt-Ingram Cancer Center

    Disclosures: Arteaga reports no relevant financial disclosures.

    Perspective
    Matthew P. Goetz

    Matthew P. Goetz

    The first report of ESR1 mutations was published more than 20 years ago. However, most of the subsequent early studies seeking to determine the clinical relevance of these alterations evaluated non-metastatic, surgically resected ER-positive tumors, and the frequency of ESR1 mutations in this setting was reported to be very low, and thus thought not to be clinically relevant.

    However, in the last 2 years, there have been multiple independent groups that have identified ESR1 mutations in the ligand-binding domain of the ER. When these mutations occur, they confer ligand independent growth, which means that ER-positive breast cancer cells with these alterations can grow in the absence of estrogen and additionally, are more resistant to anti-estrogens and aromatase inhibitors.

    In the context of the BOLERO-2 clinical trial, researchers looked for the presence of ESR1 alterations using circulating DNA, and they confirmed that ESR1 mutations are frequent (approaching 30%), and that the presence of these mutations was associated with worse prognosis.

    So, excitingly, I believe we have identified a clear mechanism for why some ER-positive breast cancers develop resistance to endocrine therapy, and there are several new classes of drugs that are being developed that may overcome this resistance. One particular class of drugs that may have activity in this setting are referred to as selective estrogen receptor down-regulators (SERD’s) and these drugs result in substantial ER degradation, and thus may have activity in patients with ESR1 mutations.  Additionally, Mayo Clinic, in collaboration with the NCI, is developing the drug endoxifen, a potent SERM with preclinical activity that is greater than tamoxifen related in part to the fact that endoxifen binds the ER with up to 100-fold greater affinity compared with tamoxifen. At this meeting, we reported the final results of a first in-human study of endoxifen and demonstrated very promising antitumor activity in aromatase inhibitor- and fulvestrant (Faslodex, AstraZeneca)-refractory metastatic breast cancer, as well as in patients with prior progression on tamoxifen.  Studies are underway to assess the activity of this drug in women with ESR1 mutations.

    Reference:

    Goetz MP, et al. Abstract PD2-03. Presented at: San Antonio Breast Cancer Symposium; Dec. 8-12, 2015; San Antonio.

    • Matthew P. Goetz, MD
    • Mayo Clinic

    Disclosures: Goetz reports no relevant financial disclosures.

    See more from San Antonio Breast Cancer Symposium