Evolving Role of the Estrogen Receptor as a Predictive Biomarker: ESR1 Mutational Status and Endocrine Resistance in Breast Cancer.

Abstract

Approximately 70% of breast cancers express the estrogen receptor alpha (ER), and endocrine therapies targeting ER are a cornerstone of their treatment. Unfortunately, a subset of patients treated with adjuvant endocrine therapy will develop metastatic disease, and patients treated with endocrine therapies in the metastatic setting will, in time, confront tumor endocrine resistance. Many mechanisms of endocrine resistance have been identified in preclinical models and clinical studies, including growth factor receptor signaling pathways (such as human epidermal growth factor receptor 2) and alterations in transcriptional coregulators. An activating mutation in ESR1, the gene encoding ER, was first described in a metastatic breast cancer in 1997. However, subsequent studies performed in primary breast cancers did not identify frequent ESR1 mutations, and the potential clinical significance of ESR1 mutations was underappreciated. It was not until 2013 that a series of studies using next-generation DNA sequencing renewed interest by demonstrating a high prevalence (11% to 55%) of ESR1 mutations in metastatic ER-positive breast cancers with prior aromatase inhibitor (AI) therapy, but not in primary breast cancers (, 1%). Most mutations occur in hotspots in the ligand-binding domain and result in constitutive, ligandindependent activity of ER, explaining how these mutations seem to be selected by the low-estrogen environment with AI therapy. Preclinical studies have shown reduced sensitivity of mutant ERs to tamoxifen, suggesting that tamoxifen might only have efficacy against ESR1mutant tumors at higher-than-standard dosages, if at all. The selective ER degrader fulvestrant seems to retain activity against mutant ERs, albeit perhaps with reduced potency, as does the cyclin-dependent kinase-4/6 inhibitor palbociclib. Altogether, these data suggest that ESR1 mutations could be a major clinical mechanism of endocrine resistance. Therefore, the clinical responsiveness of ESR1 mutant breast cancers to various endocrine therapies is now a question of great interest. Numerous recent reports have demonstrated the detection of mutant DNA alleles as tumor-specific biomarkers in cell-free DNA (cfDNA) from blood. Droplet digital polymerase chain reaction (ddPCR) is a highly sensitive and specific technique that has been used for ESR1 mutation detection in plasma. A retrospective single-institution analysis documented that ESR1mutations detected in plasma cfDNA by ddPCR were associated with a lack of response to subsequent AI therapy. However, prospective data have been lacking, and the clinical impact of alternative endocrine therapies, including fulvestrant, has not been examined. In the report that accompanies this editorial in Journal of Clinical Oncology, Fribbens et al performed a prospective-retrospective analysis of ESR1 mutations in baseline plasma samples from two randomized phase III clinical trials comparing different endocrine therapies formetastatic ER-positive breast cancer after nonsteroidal AI (NSAI). In the Study of Faslodex With or Without Concomitant Arimidex Versus Exemestane Following Progression onNon-Steroidal Aromatase Inhibitors (SoFEA) trial, women whose cancer had progressed after a period of sensitivity to NSAI, defined as recurrence after at least 12 months of adjuvant NSAI or disease progression after at least 6 months of first-line metastatic treatment with an NSAI, were randomly assigned to receive the steroidal AI exemestane, fulvestrant 250 mg, or the combination of the NSAI anastrozole and fulvestrant 250 mg. In the Palbociclib Ongoing Trials in the Management of Breast Cancer (PALOMA)-3 trial, women whose cancer had recurred within 12 months of completion of adjuvant endocrine therapy or had disease progression during palliative endocrine therapy were randomly assigned to receive fulvestrant 500 mg plus placebo versus fulvestrant 500 mg plus palbociclib. Plasma samples were available for 521 of the PALOMA-3 patients (69.1%), but only 162 of the SoFEA patients (22.4%) because most were lost in a fire. The investigators analyzed seven ESR1mutations using multiplex ddPCR, covering nearly all mutations described to date. Mutation status was analyzed as a binary outcome, with a positive mutation call requiring a minimum of two positive droplets in a minimum of 0.5 ml (PALOMA-3) or 1 ml (SoFEA) of plasma. Patients in SoFEA with ESR1 mutations (39.1%, of which 49.1% were polyclonal) had improved progression-free survival (PFS) receiving fulvestrant (n 5 45) compared with exemestane (n 5 18; hazard ratio [HR], 0.52; 95% CI, 0.30 to 0.92; P 5 .02), whereas patients with wild-type ESR1 had similar PFS receiving either treatment (HR, 1.07; 95% CI, 0.68 to 1.67; P 5 .77). In PALOMA-3, ESR1 mutations were found in the plasma of 25.3% (91/360) of patients, of which 28.6% (26/91) were polyclonal. Fulvestrant plus palbociclib improved PFS compared with fulvestrant plus placebo in patients with ESR1 mutant (HR, 0.43;