ESR1 mutation as an emerging clinical biomarker in metastatic hormone receptor-positive breast cancer

In metastatic hormone receptor-positive breast cancer, ESR1 mutations are a common cause of acquired resistance to the backbone of therapy, estrogen deprivation by aromatase inhibition. How these mutations affect tumor sensitivity to established and novel therapies are active areas of research. These therapies include estrogen receptor-targeting agents, such as selective estrogen receptor modulators, covalent antagonists, and degraders (including tamoxifen, fulvestrant, and novel agents), and combination therapies, such as endocrine therapy plus CDK4/6, PI3K, or mTORC1 inhibition. In this review, we summarize existing knowledge surrounding the mechanisms of action of ESR1 mutations and roles in resistance to aromatase inhibition. We then analyze the recent literature on how ESR1 mutations affect outcomes in estrogen receptor-targeting and combination therapies. For estrogen receptor-targeting therapies such as tamoxifen and fulvestrant, ESR1 mutations cause relative resistance in vitro but do not clearly lead to resistance in patients, making novel agents in this category promising. Regarding combination therapies, ESR1 mutations nullify any aromatase inhibitor component of the combination. Thus, combinations using endocrine alternatives to aromatase inhibition, or combinations where the non-endocrine component is efficacious as monotherapy, are still effective against ESR1 mutations. These results emphasize the importance of investigating combinatorial resistance, challenging as these efforts are. We also discuss future directions and open questions, such as studying the differences among distinct ESR1 mutations, asking how to adjust clinical decisions based on molecular surveillance testing, and developing novel therapies that are effective against ESR1 mutations.

Uncovering the differential impact of ESR1 and PIK3CA codon variants on the clinical phenotype of metastatic breast cancer (MBC) through circulating tumor DNA (ctDNA) next-generation sequencing (NGS).

1033 Background: The exposure to endocrine therapy (ET) can induce the onset of ESR1 gene alterations that have an impact on not only treatment resistance but also clinical phenotype. We previously demonstrated the potential of liquid biopsy in describing the metastatic behavior of MBC. The aim of this study was to explore the different clinical phenotype across the main ESR1 and PIK3CA codon variants. Methods: The study retrospectively analyzed a cohort of 501 MBC patients (pts) characterized for ctDNA through NGS before treatment start at Northwestern University (Chicago, IL), Massachusetts General Hospital (Boston, MA), CRO National Cancer Institute (Aviano, IT) and ASUFC Hospital (Udine, IT) between 2014 and 2020. Associations between clinical characteristics and ESR1 and PIK3CA codon variants were explored through logistic regression corrected for sites and ESR1/ PIK3CA status. Survival was tested through Cox regression both for progression-free survival (PFS) and overall survival (OS). Results: Of the total 501 pts, 289 (58%) were diagnosed with hormone-receptor positive (HRpos) MBC, 114 (23%) with HER2-positive MBC, and 93 (19%) with triple-negative MBC. ESR1 mutations were detected in 71 pts (14%) and PIK3CA in 154 pts (31%). The most represented ESR1 gene mutations were found in codons 380 (9%), 536 (23%), 537 (34%), and 538 (34%), while alterations in codons 542 (19%), 545 (21%), and 1047 (60%) were the most common for PIK3CA. As expected, ESR1 mutations were found only in HRpos pts previously exposed to ET (P < 0.001). No significant differences were observed for PIK3CA. After multivariable analysis, ESR1mutations were confirmed as highly associated with liver and bone metastases (OR 3.31, P < 0.001 and OR 5.09, P < 0.001). Moreover, an association with lung (OR 2.07, P = 0.010) was observed in this cohort. After multivariable analysis, codon 537 mutations were associated with bone involvement (OR 12.97, P = 0.014), codon 538 with liver (OR 4.73, P = 0.010), and codon 536 with soft tissue (OR 5.84, P = 0.006) and liver (OR 4.06, P = 0.048). PIK3CA mutations were associated with bone (OR 2.61, P < 0.001) and lung metastases (OR 1.62, P = 0.044). Specifically, codon 1047 mutations were the primary driver (OR 3.14, P = 0.001 and OR 1.97, P = 0.019). In HRpos MBC, baseline mutations in ESR1 codon 537 and 538 had a negative impact on OS (HR 3.73, P < 0.010 and HR 2.99, P < 0.021), while 380 and 536 had a negative impact on PFS (HR 18.98, P < 0.001 and HR 2.60, P = 0.015). No impact was observed across PIK3CA gene variants. Conclusions: This study showed the different tumor biology across ESR1 and PIK3CA gene variants. As novel selective estrogen receptor degraders (SERDS) and PIK3CA inhibitors are gaining momentum as new ET options in MBC, these results highlight the future pivotal role of ctDNA NGS in refining tumor biology characterization.

acelERA Breast Cancer (BC): Phase II study evaluating efficacy and safety of giredestrant (GDC-9545) versus physician’s choice of endocrine monotherapy in patients (pts) with estrogen receptor-positive, HER2-negative (ER+/HER2-) locally advanced or metastatic breast cancer (LA/mBC).

TPS1100 Background: The standard-of-care therapy for ER+ BC typically involves modulation of estrogen synthesis and/or ER activity. Despite disease progression with standard treatments, growth and survival of the majority of tumors are thought to remain dependent on ER signaling; therefore, pts with ER+ BC can still respond to second- or third-line endocrine treatment (ET) after progression on prior therapy (Di Leo 2010; Baselga 2012). ESR1 mutations may drive estrogen-independent transcription and proliferation leading to resistance. The highly potent, nonsteroidal oral selective ER degrader, giredestrant, achieves robust ER occupancy and is active regardless of ESR1 mutation status. Phase I data have shown that giredestrant is well tolerated and active both as a single agent and in combination with the cyclin-dependent kinase 4/6 inhibitor (CDK4/6i), palbociclib (Lim 2020). Single-agent giredestrant has also shown encouraging antitumor activity in pts previously treated with fulvestrant and/or a CDK4/6i. Methods: acelERA BC (NCT04576455) is a randomized, open-label, multicenter phase II study evaluating the efficacy and safety of giredestrant vs. physician’s choice of ET (fulvestrant or aromatase inhibitor) in males, or postmenopausal or pre/perimenopausal females with ER+/HER2– LA/mBC who have received 1–2 prior lines of systemic therapy in the LA or mBC settings, at least one of which must be ET. Randomization: 1:1 to receive giredestrant (30 mg PO QD on Days 1–28 of each 28-day cycle) or physician’s choice of ET per local guidelines. Men and pre-/perimenopausal women will receive a luteinizing hormone-releasing hormone agonist. Eligibility: ≥18 years, ECOG PS 0–1, histologically or cytologically confirmed diagnosis of LA (recurrent or progressed) or metastatic adenocarcinoma of the breast, measurable disease (per modified RECIST v1.1) or evaluable bone lesions, and ER+/HER2– tumors (locally assessed). Primary endpoint: progression-free survival (PFS; investigator-assessed per RECIST v1.1). Secondary endpoints: overall survival, objective response rate, duration of response, clinical benefit rate, PFS in pts with baseline ESR1 mutations, and quality of life. Safety, pharmacokinetics, biomarkers, and health status utility will also be assessed. Stratification: site of disease (visceral vs. nonvisceral), prior treatment with CDK4/6i (yes vs. no), and prior treatment with fulvestrant (yes vs. no). PFS will be compared using a stratified log-rank test; median PFS, using Kaplan–Meier analyses. Recruitment for the global enrollment phase is ongoing, the first patient was enrolled November 27, 2020. Clinical trial information: NCT04576455 .

Safety and activity of single-agent giredestrant (GDC-9545) from a phase Ia/b study in patients (pts) with estrogen receptor-positive (ER+), HER2-negative locally advanced/metastatic breast cancer (LA/mBC).

1017 Background: Targeting ER activity and/or E synthesis is a mainstay of ER+ BC treatment, but many pts relapse during/after adjuvant endocrine therapy (ET) or develop resistance via ESR1 mutations that drive E-independent transcription and proliferation. Most tumors remain ER signaling-dependent and pts may respond to second-/third-line ET after disease progression (PD) on prior therapies (Di Leo 2010; Baselga 2012). Giredestrant, a highly potent, nonsteroidal oral selective ER degrader, achieves robust ER occupancy, is active despite ESR1 mutations, and was well tolerated ± palbociclib with encouraging antitumor activity in the nonrandomized, open-label, dose-escalation and -expansion, phase Ia/b GO39932 study (NCT03332797; Jhaveri 2019; Lim 2020). We present updated interim data from the dose-escalation and -expansion single-agent giredestrant cohorts. Methods: Pts had ≤2 prior therapies in the LA/mBC setting with disease recurrence/PD while being treated with adjuvant ET for ≥24 mo and/or ET in the LA/mBC setting, and derived a clinical benefit (CB) from therapy (tumor response/stable disease [SD] ≥6 mo). Pts received 10, 30, 90/100, or 250 mg PO giredestrant QD on D1–28 of each 28-day cycle. Pts were postmenopausal (medical menopause on LHRH agonists was allowed with ≥100 mg giredestrant). Results: Clinical cutoff: Jul 31, 2020; median prior therapy lines in the LA/mBC setting: 1; mean dose intensity: 98%. Safety/activity: see the table below. No adverse events (AEs) led to study drug withdrawal. No dose-limiting toxicities (DLTs) occurred; maximum tolerated dose was not reached. Most common AEs in 107 pts: fatigue (22; 21%), arthralgia (18; 17%), and nausea (17; 16%); largely grade 1/2. Related grade 3 AEs were infrequent (5; 5%); none were grade 4/5 per investigator assessment (grade 5 duodenal perforation occurred with 90/100 mg after stopping giredestrant due to PD). 8 (7%) had bradycardia (none with 10 mg or the 30 mg phase 3 dose; all grade 1 except one grade 2 at 250 mg; no treatment interruptions/dose reductions were required). Objective responses and CB were observed at all doses. Conclusions: Single-agent giredestrant was well tolerated at all doses, with no DLTs. AEs were generally low grade and in keeping with expected AEs for ETs. Clinical activity was observed at all doses. Updated data, including biomarker and correlative data, will be presented. Clinical trial information: NCT03332797 .[Table: see text]

Phase I/II study of H3B-6545, a novel selective estrogen receptor covalent antagonist (SERCA), in estrogen receptor positive (ER+), human epidermal growth factor receptor 2 negative (HER2-) advanced breast cancer.

1018 Background: H3B-6545, a selective, small molecule covalent antagonist of ERα demonstrated preclinical and preliminary clinical activity against ER+ breast cancer (Hamilton EP, SABCS, 2020). This study evaluated the activity and tolerability of H3B-6545 in patients (pts) with metastatic ER+, HER2-, breast cancer refractory to endocrine therapy. Methods: Patients received H3B-6545 once daily at the recommended phase II dose of 450 mg. The primary objective of the phase II is to estimate the objective response rate (ORR), progression-free survival (PFS), clinical benefit rate (CBR) and secondary objectives include safety. Results: 83 pts were treated with 450 mg in the phase II part of the trial. Additionally, 11 pts were treated with 450 mg in the phase I part of the trial and are included in this analysis. Median age was 62 years (range: 38 to 87 years), 81% had liver and/or lung metastases, and the median number of prior therapies for metastatic disease was 3 (range: 1 to 8). Prior CDK4/6 inhibitors, aromatase inhibitors, fulvestrant, and chemotherapy were received by 85%, 80%, 72%, and 50% of the pts, respectively. 58 pts (62%) had detectable ESR1 mutations in liquid biopsies, including 10 (11%) and 19 pts (20%) who had clonal Y537S and clonal D538G mutation, respectively. As of January 29, 2021, grade (gr) 2 or higher adverse events (AE) reported in ≥10% were anemia (19%), fatigue (16%), nausea (17%), and diarrhea (12%). Laboratory gr 2 or higher abnormalities reported in ≥10% pts were creatinine clearance decrease (38%), hemoglobin decrease (37%), bilirubin increase (12%), ALT increase (14%), AST increase (13%), and creatinine increase (11%). AE of gr 1 sinus bradycardia (asymptomatic) was reported in 34% and gr 2 (symptomatic, no intervention needed) was reported in 5%. Gr 2 and 3 QTcF prolongation were reported in 2 and 3 pts, respectively. There were no treatment-related deaths. Efficacy estimates are presented in the table below. Responses were observed in heavily pretreated pts, pts with visceral metastases and in pts who received prior fulvestrant, CDK4/6 inhibitor, and/or chemotherapy in the metastatic setting. Conclusions: H3B-6545 has a manageable safety profile and demonstrated single-agent anti-tumor activity in heavily pretreated ER+, HER2- mBC patients. Clinical activity was observed in pts with ESR1 mutations. Clinical trial information: NCT03250676 .[Table: see text]

ESR1 NAPA Assay: Development and Analytical Validation of a Highly Sensitive and Specific Blood-Based Assay for the Detection of ESR1 Mutations in Liquid Biopsies

A considerable number of estrogen receptor-positive breast cancer (ER+ BrCa) patients develop resistance to endocrine treatment. One of the most important resistance mechanisms is the presence of ESR1 mutations. We developed and analytically validated a highly sensitive and specific NaME-PrO-assisted ARMS (NAPA) assay for the detection of four ESR1 mutations (Y537S, Y537C, Y537N and D538G) in circulating tumour cells (CTCs) and paired plasma circulating tumour DNA (ctDNA) in patients with ER+ BrCa. The analytical specificity, analytical sensitivity and reproducibility of the assay were validated using synthetic oligos standards. We further applied the developed ESR1 NAPA assay in 13 ER+ BrCa primary tumour tissues, 13 non-cancerous breast tissues (mammoplasties) and 64 liquid biopsy samples: 32 EpCAM-positive cell fractions and 32 paired plasma ctDNA samples obtained at different time points from 8 ER+ metastatic breast cancer patients, during a 5-year follow-up period. Peripheral blood from 11 healthy donors (HD) was used as a control. The developed assay is highly sensitive (a detection of mutation-allelic-frequency (MAF) of 0.5% for D538G and 0.1% for Y537S, Y537C, Y537N), and highly specific (0/13 mammoplasties and 0/11 HD for all mutations). In the plasma ctDNA, ESR1 mutations were not identified at the baseline, whereas the D538G mutation was detected in five sequential ctDNA samples during the follow-up period in the same patient. In the EpCAM-isolated cell fractions, only the Y537C mutation was detected in one patient sample at the baseline. A direct comparison of the ESR1 NAPA assay with the drop-off ddPCR using 32 identical plasma ctDNA samples gave a concordance of 90.6%. We present a low cost, highly specific, sensitive and robust assay for blood-based ESR1 profiling. The clinical performance of the ESR1 NAPA assay will be prospectively evaluated in a large number of well-characterized patient cohorts.