AMEERA-3: Randomized Phase II Study of Amcenestrant (Oral Selective Estrogen Receptor Degrader) Versus Standard Endocrine Monotherapy in Estrogen Receptor–Positive, Human Epidermal Growth Factor Receptor 2–Negative Advanced Breast Cancer

Abstract

PURPOSE

Amcenestrant (oral selective estrogen receptor degrader) demonstrated promising safety and efficacy in earlier clinical studies for endocrine-resistant, estrogen receptor–positive/human epidermal growth factor receptor 2–negative (ER+/HER2–) advanced breast cancer (aBC).

PATIENTS AND METHODS

In AMEERA-3 (ClinicalTrials.gov identifier: NCT04059484), an open-label, worldwide phase II trial, patients with ER+/HER2– aBC who progressed in the (neo)adjuvant or advanced settings after not more than two previous lines of endocrine therapy (ET) were randomly assigned 1:1 to amcenestrant or single-agent endocrine treatment of physician's choice (TPC), stratified by the presence/absence of visceral metastases, previous/no treatment with cyclin-dependent kinase 4/6 inhibitor, and Eastern Cooperative Oncology Group performance status (0/1). The primary end point was progression-free survival (PFS) by independent central review, compared using a stratified log-rank test (one-sided type I error rate of 2.5%).

RESULTS

Between October 22, 2019, and February 15, 2021, 290 patients were randomly assigned to amcenestrant (n = 143) or TPC (n = 147). PFS was numerically similar between amcenestrant and TPC (median PFS [mPFS], 3.6 v 3.7 months; stratified hazard ratio [HR], 1.051 [95% CI, 0.789 to 1.4]; one-sided P = .643). Among patients with baseline mutated ESR1; (n = 120 of 280), amcenestrant numerically prolonged PFS versus TPC (mPFS, 3.7 v 2.0 months; stratified HR, 0.9 [95% CI, 0.565 to 1.435]). Overall survival data were immature but numerically similar between groups (HR, 0.913; 95% CI, 0.595 to 1.403). In amcenestrant versus TPC groups, treatment-emergent adverse events (any grade) occurred in 82.5% versus 76.2% of patients and grade ≥3 events occurred in 21.7% versus 15.6%.

CONCLUSION

AMEERA-3 did not meet its primary objective of improved PFS with amcenestrant versus TPC although a numerical improvement in PFS was observed in patients with baseline ESR1 mutation. Efficacy and safety with amcenestrant were consistent with the standard of care for second-/third-line ET for ER+/HER2– aBC.

INTRODUCTION

As of 2020, female breast cancer is the most commonly diagnosed cancer globally,¹ with the majority (68%-73%) of women presenting with hormone receptor–positive (estrogen receptor–positive [ER+] and/or progesterone receptor [PgR]–positive), human epidermal growth factor receptor 2–negative (HER2–) disease.^2,3 For ER+/HER2– advanced breast cancer (aBC), endocrine therapy (ET) with an aromatase inhibitor (AI) or the selective ER degrader (SERD) fulvestrant in combination with a cyclin-dependent kinase 4/6 inhibitor (CDK4/6i) is the standard-of-care (SOC) treatment in the first-line setting; single-agent ET alone or in combination with targeted agents is recommended for treatment in the second line, depending on previous treatment regimens.⁴

CONTEXT

• Key Objective

• Does the oral selective estrogen receptor degrader (SERD) amcenestrant improve progression-free survival (PFS) compared with standard endocrine monotherapy in patients with pretreated estrogen receptor–positive/human epidermal growth factor receptor 2–negative advanced breast cancer (aBC)?

• Knowledge Generated

• In a randomized phase II trial, amcenestrant did not lead to an improvement in PFS compared with standard endocrine therapy although there was a trend toward improved PFS in patients with tumors with baseline ESR1 mutation. These results suggest that there is no additional benefit of amcenestrant monotherapy compared with current standard of care in an unselected population in this setting, but are consistent with other data in aBC, suggesting that oral SERDs seem to have larger benefits in patients with tumors harboring ESR1 mutations.

• Relevance (K.D. Miller)

• This is a negative trial, amcenestrant did not improve PFS overall or in any defined patient subset in this randomized phase II trial. While cross trial comparisons are always difficult, these results suggest that not all oral SERDS are equally effective.*

  *Relevance section written by JCO Senior Deputy Editor Kathy D. Miller, MD.

However, acquired resistance eventually occurs in 30%-40% of ER+ tumors over the course of treatment,⁵ often mediated by mutations in the ESR1 gene after AI therapy and genomic aberrations in non–ER-dependent genes.⁶ Tumors that develop resistance to first-line ETs but remain dependent on ER signaling frequently respond to an alternative ET alone or in combination with targeted therapy.⁷

SERDs are competitive ER antagonists with potent dual activity that antagonizes and degrades the ER, resulting in inhibition of the ER signaling pathway.⁸ Fulvestrant⁹ has demonstrated antitumor activity as first-line monotherapy in endocrine-naïve patients¹⁰ and in combination with targeted agents as second-line therapy in endocrine-resistant patients,^11-13 including in patients with ESR1 mutations.¹⁴ However, fulvestrant has pharmacokinetic limitations (poor oral bioavailability requiring large-volume intramuscular injections,¹⁵ with up to 1 month to reach the steady state^16,17) and pharmacodynamic limitations (suboptimal ER occupancy with the maximum 500-mg once monthly dose), which have been associated with early disease progression¹⁸ and acquired resistance to treatment, especially in the presence of the ESR1-Y537S mutation.¹⁹ The development of novel oral SERDs with improved oral bioavailability, ER occupancy, and antitumor activity may address these limitations.

In a phase I/II trial (AMEERA-1; ClinicalTrials.gov identifier: NCT03284957) in postmenopausal women with ER+/HER2– aBC, most of whom were heavily pretreated, monotherapy with the oral SERD amcenestrant demonstrated encouraging antitumor activity irrespective of ESR1 mutation status and a favorable safety profile.²⁰ Here, we report the primary analysis of AMEERA-3 (ClinicalTrials.gov identifier: NCT04059484), a phase II trial comparing the efficacy and safety of amcenestrant monotherapy versus single-agent endocrine treatment of physician's choice (TPC) in patients with ER+/HER2– aBC that had progressed on or after ET.

PATIENTS AND METHODS

Study Design and Patients

AMEERA-3 was an international, prospective, open-label, phase II trial in which patients were randomly assigned 1:1 to receive amcenestrant 400 mg once daily or TPC, with a block random assignment size of 4 (Data Supplement [Fig S1], online only). The potential TPC was selected before random assignment, depending on the patient's medical condition and in accordance with the approved label: fulvestrant 500 mg, administered on days 1 and 15 of cycle 1, and at day 1 of each 28-day cycle thereafter⁹; oral AI once daily with anastrozole 1 mg, letrozole 2.5 mg, or exemestane 25 mg; or the selective ER modulator tamoxifen 20 mg once daily. Random assignment was stratified according to the absence/presence of visceral metastases (≥1 liver or lung metastasis), previous treatment with CDK4/6i (yes/no), and an Eastern Cooperative Oncology Group performance status of 0 or 1.

Key inclusion criteria included the following: women (any menopausal status; the postmenopausal age threshold was 60 years and older) and men 18 years and older (pre-/perimenopausal women and men with no previous bilateral orchiectomy treated with gonadotropin-releasing hormone agonist ≥4 weeks before random assignment and during study treatment), documentation of ER+ (≥1% positive stained cells) and HER2– status on the most recent archival tissue, evidence of locally advanced or metastatic disease, ≤1 previous chemotherapy or targeted therapy for aBC, pretreatment with CDK4/6i in countries where approved (CDK4/6i-naïve patients limited to ≤20%), progression after ≥6 months of continuous ET or relapse while on adjuvant ET but after the first 2 years or within 12 months of completing adjuvant ET, and ≤2 previous lines of ET. For a complete list of criteria, see the associated protocol. The Protocol (online only) was approved by the institutional review board or independent ethics committee at each site and complied with the International Ethical Guidelines for Biomedical Research Involving Human Subjects, Good Clinical Practice guidelines, Declaration of Helsinki, and local laws. All patients provided written informed consent. An independent data monitoring committee (Data Supplement [Table S1]) reviewed the safety data and progress of the study.

Prespecified End Points

The primary end point was progression-free survival (PFS), defined as the time from random assignment to disease progression or any-cause death, as assessed by independent central review (ICR) per RECIST version 1.1.²¹ The key secondary end point was overall survival (OS), defined as the time from random assignment to death from any cause.

Other secondary end points included the following: objective response rate (ORR), disease control rate, clinical benefit rate (CBR), duration of response, PFS by baseline ESR1 mutation status, pharmacokinetics, health-related quality of life (HRQOL), and safety, including treatment-emergent adverse events (TEAEs), serious adverse events (SAEs), and laboratory abnormalities.

Tertiary/exploratory end points included ESR1 mutation analysis, evaluation of tumor biomarkers, and gene expression profiles.

Statistical Analysis

We estimated that 201 ICR-assessed PFS events in the two arms would provide approximately 85% power at a one-sided 2.5% significance level at a hazard ratio (HR) of 0.65, accounting for the futility interim analysis at 50% of targeted events.

PFS was compared between groups using a stratified log-rank test and a one-sided type I error rate of 2.5%. The Kaplan-Meier method was used to summarize PFS and to estimate the median PFS (mPFS) with associated 95% two-sided CIs. HR estimates and corresponding 95% two-sided CIs were calculated using a stratified Cox model. Prespecified subgroup analyses of PFS were performed on the basis of HRs and corresponding 95% two-sided CIs with the unstratified Cox model. For details on other analyses, see the Data Supplement.

RESULTS

Between October 22, 2019, and February 15, 2021, 290 patients from 88 study locations across 22 countries were randomly assigned to receive amcenestrant (143 patients) or TPC (147 patients: fulvestrant, n = 132 [89.8%]; AI, n = 10 [6.8%]; tamoxifen, n = 5 [3.4%]). Baseline characteristics of the intent-to-treat (ITT) population were well balanced between treatments (Table 1).

TABLE 1.

Baseline Demographic and ICR-Assessed Disease Characteristics (ITT population)

At the data cutoff for the final PFS analysis (February 15, 2022), treatment was ongoing in 20 (14.0%) amcenestrant recipients and 20 (13.6%) TPC recipients. The median duration of exposure to amcenestrant was 16 weeks, and that to TPC was 15.9 weeks, with 56 (39.2%) and 54 (36.7%) patients receiving treatment for ≥24 weeks, respectively. The median relative dose was 100.0% in both groups, and the proportion of patients having a relative dose between 80% and 100% was 95.8% for amcenestrant and 97.3% for TPC. Treatment discontinuation because of disease progression occurred in 112 (78.3%) amcenestrant recipients and 112 (76.2%) TPC recipients (Fig 1). Seven (2.4%) patients discontinued treatment because of adverse events: five in the amcenestrant arm (grade 1-2: anemia, angina pectoris/cardiac failure/mitral valve disease, laryngeal edema, dyspepsia; grade 3: rash) and two in the TPC arm (grade 3: COVID-19; grade 5: pneumonia). Dyspepsia and rash were related to amcenestrant treatment.

FIG 1.

CONSORT diagram. TPC, single-agent endocrine treatment of physician's choice.

Efficacy

The trial did not meet its primary objective in the final PFS analysis as amcenestrant was not statistically superior to TPC. At a median follow-up of 11.2 months (IQR, 5.7-16.5), 195 ICR-assessed PFS events had occurred (amcenestrant, n = 100; TPC, n = 95). The mPFS was 3.6 months (95% CI, 2.0 to 3.9) with amcenestrant and 3.7 months (95% CI, 2.0 to 4.9) with TPC (HR, 1.051 [95% CI, 0.789 to 1.4]; one-sided P = .6437; Fig 2A). The PFS rates at 6 and 12 months were 35.5% (95% CI, 27.2 to 43.9) and 20.4% (95% CI, 13.4 to 28.4) in the amcenestrant arm and 37.6% (95% CI, 29.2 to 46.0) and 24.6% (95% CI, 16.6 to 33.3) in the TPC arm, respectively.

FIG 2.

Kaplan-Meier analysis of PFS on the basis of ICR assessment in (A) the ITT population or (B) patients with baseline mutated ESR1. The one-sided P value was computed from a stratified log-rank test according to stratification factors (presence of visceral metastasis, previous treatment with CDK4/6i, and ECOG PS). HR and CIs were computed from a stratified Cox model according to stratification factors (presence of visceral metastasis, previous treatment with CDK4/6i, and ECOG PS). CDK4/6i, cyclin-dependent kinase 4/6 inhibitor; ECOG PS, Eastern Cooperative Oncology Group performance status; HR, hazard ratio; ICR, independent central review; ITT, intent-to-treat; mPFS, median progression-free survival; PFS, progression-free survival; TPC, single-agent endocrine treatment of physician's choice.

The primary analysis findings were supported by investigator-assessed PFS (HR, 0.944 [95% CI, 0.725 to 1.228]; Data Supplement [Fig S2]). ICR-assessed analyses of PFS according to baseline characteristics were consistent with the primary analysis (Fig 3). Baseline ESR1 mutations were present in 65 of 140 (46.4%) and 55 of 140 (39.3%) patients in the amcenestrant and TPC arms, respectively. Among patients with mutated ESR1 (mESR1), the mPFS was 3.7 months (95% CI, 1.9 to 7.2) with amcenestrant versus 2.0 months (95% CI, 1.9 to 4.3) with TPC (stratified HR, 0.9 [95% CI, 0.565 to 1.435]; Fig 2B). In patients with wild-type ESR1 (wt-ESR1), the mPFS was 3.5 months (95% CI, 2.0 to 3.7) with amcenestrant versus 3.9 months (95% CI, 3.6 to 9.2) with TPC (stratified HR, 1.305 [95% CI, 0.88 to 1.937]).

FIG 3.

Subgroup analysis of PFS on the basis of ICR assessment by stratification factors per IRT and demographic/baseline characteristics. ^aThe HR estimates and corresponding 95% two-sided CIs used an unstratified Cox proportional hazard model. ^bDefined as at least one liver or lung metastasis. ^cOn the basis of ICR assessment at baseline. ^dThe postmenopausal age threshold was 60 years and older, and data are among female patients only. CDK4/6i, cyclin-dependent kinase 4/6 inhibitor; ECOG PS, Eastern Cooperative Oncology Group performance status; HR, hazard ratio; ICR, independent central review; IRT, interactive response technology; PFS, progression-free survival; TPC, single-agent endocrine treatment of physician's choice.

At data cutoff, 40 deaths in the amcenestrant arm and 46 deaths in the control arm had occurred. OS analysis (data not matured) indicated a numerically similar survival trend between groups (HR, 0.913 [95% CI, 0.595 to 1.403]). Tumor response–based end points are reported in Table 2 (ICR-assessed) and the Data Supplement ([Table S2]; investigator-assessed sensitivity analysis).

TABLE 2.

ICR-Assessed Antitumor Activity of Amcenestrant Compared With TPC Groups

Biomarker Analysis

Among patients with the ESR1-Y537S mutation (amcenestrant, n = 27 of 140 [19.3%]; TPC, n = 26 of 141 [18.4%]), numerically longer PFS was observed with amcenestrant versus TPC (mPFS, 3.8 v 1.9 months; stratified HR, 0.431 [95% CI, 0.205 to 0.909]; Data Supplement [Fig S3]), with a trend of better ORR and CBR (Data Supplement [Table S3]). In patients with ESR1 data both at baseline and on treatment (amcenestrant, n = 90; TPC, n = 82), 57.7% of ESR1 mutations detected at baseline were cleared on treatment with amcenestrant versus 31.3% for TPC (Data Supplement [Fig S4]). Baseline low Ki67 expression and high ER and PgR H-scores were each associated with numerically longer PFS in both groups, and a trend toward longer PFS was observed for high ER activity score with amcenestrant versus TPC (Data Supplement [Table S4]).

Safety

In the safety population (amcenestrant, n = 143; TPC, n = 147), any-grade TEAEs occurred in 118 of 143 (82.5%) patients in the amcenestrant arm and 112 of 147 (76.2%) in the TPC arm, mostly grade 1-2. The most frequent TEAEs are reported in Table 3. SAEs occurred in 23 of 143 (16.1%) amcenestrant recipients and 15 of 147 (10.2%) TPC recipients. One SAE in each group was treatment related (grade 3 gastritis with amcenestrant; grade 3 angina pectoris with TPC). Amcenestrant treatment–related AEs of eye pain, blurred vision, and photopsia occurred (one patient each), but all were grade 1 events. No clinically significant treatment-related cardiac or ocular safety findings were observed with amcenestrant. During the on-treatment period (from administration of the first dose of study drug to 30 days after the last dose), one TEAE not related to progressive disease resulting in death occurred: pneumonia with fulvestrant in the TPC arm (not treatment related).

TABLE 3.

Treatment-Emergent Adverse Events Occurring in ≥10% of Patients in Either Treatment Arm by Grade

HRQOL

Proportionally fewer patients treated with amcenestrant versus TPC experienced sustained deterioration in pain (37.4% v 45.3%). The median time to sustained deterioration in pain improved with amcenestrant versus TPC (12.1 months [95% CI, 7.5 to 17.8] v 8.3 months [95% CI, 4.6 to 16.6]), corresponding to a 20% risk reduction (HR, 0.80 [95% CI, 0.54 to 1.18; Data Supplement [Fig S5]). Results were similar between treatments for physical functioning, role functioning, and HRQOL (Data Supplement [Figs S6 and S7]).

DISCUSSION

The open-label, randomized, phase II AMEERA-3 trial did not meet its primary objective as amcenestrant monotherapy did not significantly extend PFS compared with TPC in patients with ER+/HER2– aBC. Patients treated with amcenestrant achieved numerically similar mPFS compared with TPC (3.6 v 3.7 months) and 6-month (35.5% v 37.6%) and 12-month (20.4% v 24.6%) PFS rates. To minimize bias and variability in determining efficacy, the PFS end point was based on ICR; however, the analysis by local investigators was consistent with the PFS assessment by ICR. Subgroup analysis of PFS generally supported the primary results. On the basis of a hierarchical testing strategy, the key secondary end point of OS was reported with descriptive statistics in an interim analysis. An HR of 0.913 (95% CI, 0.595 to 1.403) indicated a numerically similar survival trend between arms. Because of the study outcome of the primary objective, no further analysis of OS in AMEERA-3 will be conducted.

Most patients in the amcenestrant arm had received at least four treatment cycles (16 weeks), and the safety profile was manageable, with mostly mild or moderate (grade 1 or 2) TEAEs and no new safety signals. Grade ≥3 TEAEs occurred with similar frequency between the amcenestrant and TPC arms (21.7% v 15.6%). The most frequently reported amcenestrant TEAE was nausea, which occurred in 29 of 143 (20.3%) patients. Of these, 20 of 29 (69.0%) were related to amcenestrant treatment (all grade 1-2) and 3 of 29 (10.3%) were grade 3 (and not related to treatment). Two patients discontinued treatment because of amcenestrant treatment–related dyspepsia and rash, and symptoms eventually resolved. Clinically significant cardiac or ocular safety findings, which have been associated with other SERDs in development,^22-25 did not occur. Overall, amcenestrant maintained quality of life and performed numerically better than TPC in some patient-reported outcomes measuring pain.

Among the four randomized phase II/III trials comparing oral SERD monotherapy versus SOC ET monotherapy in ER+/HER2– aBC, two trials (EMERALD²⁶ and SERENA-2²⁵) demonstrated significant improvements in PFS, the first with elacestrant (HR, 0.70 [95% CI, 0.55 to 0.88]) and the second with camizestrant (75 mg once daily: HR, 0.58 [90% CI, 0.41 to 0.81]; 150 mg once daily: HR, 0.67 [90% CI, 0.48 to 0.92]), whereas statistical significance on PFS was not reached in acelERA (giredestrant; HR, 0.81 [95% CI, 0.60 to 1.10])²⁷ and AMEERA-3 (amcenestrant: HR, 1.051 [95% CI, 0.789 to 1.4]). The absolute mPFS (3.6 months) observed with amcenestrant in AMEERA-3 was comparable with the mPFS (2.8 months) observed with elacestrant in the EMERALD trial²⁶ but lower than the mPFS observed with camizestrant 75 mg once daily (7.2 months) and 150 mg once daily (7.7 months) in SERENA-2²⁵ and with giredestrant (5.6 months) in acelERA.²⁷ However, indirect cross-trial comparisons should be cautioned because of heterogeneity in the patient populations between studies. Previous studies have shown PFS superiority of single-agent fulvestrant compared with AI²⁸ or tamoxifen²⁹ monotherapy. In AMEERA-3, patients in the TPC arm were treated with fulvestrant (89.8%), AI (6.8%), or tamoxifen (3.4%), whereas in EMERALD, fewer control patients were treated with fulvestrant (69.3%) and tamoxifen was excluded from the SOC regimen.²⁶ It is possible the high proportion of patients receiving fulvestrant in AMEERA-3 might have led to prolonged mPFS in the TPC arm. The pronounced initial drop in PFS in both arms was potentially indicative of an endocrine-resistant population, consistent with observations in EMERALD.²⁶ However, the proportion of patients receiving fulvestrant in the control arm of acelERA (75%) was closer to that of EMERALD, and no significant difference in outcomes was observed between treatments in acelERA.²⁷ In addition, 100% of the control arm in SERENA-2 received fulvestrant, with both doses of camizestrant demonstrating significant improvements in PFS versus fulvestrant.²⁵ Differences in previous treatment must also be considered; no previous fulvestrant for aBC was permitted in SERENA-2,²⁵ whereas previous fulvestrant was noted in 30.4% of patients in EMERALD,²⁶ 19% of patients in acelERA,²⁷ and 9.7% of patients in AMEERA-3. Previous lines of therapy in the advanced setting also varied, with more patients having one or two previous lines of ET in AMEERA-3 (82.1% and 11.4%, respectively) and EMERALD (56.6%; 43.4%)²⁶ or one or two previous lines of systemic therapy (≥1 must have been ET) in acelERA (71%; 28%),²⁷ whereas all patients in SERENA-2 had either 0 (31.3%) or 1 (68.8%) previous line of ET in the advanced setting.²⁵ Previous CDK4/6i treatment differed across trials, with 42% of patients in acelERA having received previous CDK4/6i, 49.6% in SERENA-2, 79.0% in AMEERA-3, and 100% in EMERALD.^25-27

Although mutations in ESR1 predict resistance to AI therapy, resulting in shorter PFS,^30,31 fulvestrant as monotherapy^32,33 or in combination with a CDK4/6i^14,34 previously demonstrated similar PFS outcomes between wt-ESR1 and mESR1 cohorts in patients with ER+/HER2– aBC who progressed on previous AI. In EMERALD, elacestrant demonstrated significantly longer PFS versus SOC in a mESR1 cohort (HR, 0.55 [95% CI, 0.39 to 0.77]), corresponding to a 45% reduction in progression or death.²⁶ Similarly, longer PFS with camizestrant versus fulvestrant in the overall population of SERENA-2 appeared to be driven by the greater effect in patients with baseline mESR1 (camizestrant 75 mg once daily: HR, 0.33 [90% CI, 0.18 to 0.58]; camizestrant 150 mg once daily: HR, 0.55 [0.33 to 0.89]) as no significant improvements in PFS were observed in patients with wt-ESR1 (camizestrant 75 mg once daily: HR, 0.78 [0.50 to 1.22]; camizestrant 150 mg once daily: HR, 0.76 [0.48 to 1.20]).²⁵ Although previous results had shown similar PFS with fulvestrant regardless of ESR1 status,^32,33 recent studies suggest mixed results regarding response to fulvestrant in patients with mESR1.^25,31,35 In SERENA-2, patients with baseline mESR1 had a shorter mPFS compared with wt-ESR1 in the fulvestrant arm (2.2 v 7.2 months).²⁵ In addition, patients in the fulvestrant arm had a higher rate of baseline ESR1 mutation (47.9%) versus camizestrant 75 mg once daily (29.7%) or 150 mg once daily (35.6%), which might have contributed to the poorer response in the fulvestrant arm, if indeed the ESR1 mutation was more resistant to fulvestrant. Results from acelERA also indicated a more pronounced numerical benefit of giredestrant versus SOC ET monotherapy in patients with baseline mESR1 (HR, 0.60; 95% CI, 0.35 to 1.03).²⁷ In AMEERA-3, among patients with baseline mESR1 (46.4%), amcenestrant showed a trend of longer PFS compared with TPC (3.7 v 2.0 months; stratified HR, 0.9 [95% CI, 0.565 to 1.435]). Taken together, these results suggest that a better treatment effect compared with SOC in patients with mESR1 may be a class effect of new-generation oral SERDs and may be driving the benefit observed in the overall populations. Indeed, the US Food and Drug Administration (FDA) recently approved elacestrant for the treatment of postmenopausal women or adult men with ER+/HER2–, ESR1-mutated aBC after disease progression after at least one line of ET,³⁶ along with approval of the companion diagnostic liquid biopsy test to identify ESR1 mutations.³⁷

The ESR1-Y537S mutation, in particular, has shown resistance to fulvestrant and AI, with shorter PFS compared with other ESR1 variants.^33,38 Preclinical data also support differential sensitivity to fulvestrant according to the ESR1 mutation type, with tumors bearing the ESR1-Y537S mutation being most resistant.¹⁹ The PFS benefit observed with oral SERDs versus SOC ET in patients with baseline mESR1 across the four trials might have also varied in part depending on the proportions of select ESR1 mutations that confer resistance to fulvestrant but still respond to potent oral SERDs. This, in turn, may affect the extent of the difference in response between oral SERDs and ET SOC and help explain conflicting results reported for response to fulvestrant in patients with mESR1.^14,25,31,32 In AMEERA-3, patients with the ESR1-Y537S mutation had numerically longer PFS with amcenestrant versus TPC, suggesting that the ESR1-Y537S mutation does not confer resistance to amcenestrant.

Previous studies have shown that addition of targeted therapy to ET improves PFS. In PALOMA-3, fulvestrant plus the CDK4/6i palbociclib more than doubled PFS compared with fulvestrant alone.¹¹ Similar results were observed in BOLERO-2³⁹ with exemestane plus the mTORi everolimus and in SOLAR-1⁴⁰ with fulvestrant plus the phosphoinositide 3-kinase inhibitor alpelisib in patients with PIK3CA-mutated aBC who relapsed or progressed on previous AI monotherapy. AMEERA-3 did not evaluate combination therapies because it was designed to assess amcenestrant as a backbone ET for aBC in the second-/third-line setting, and most patients were previously treated with CDK4/6i, consistent with current practice guidelines. However, parts of the AMEERA-1 study assessed amcenestrant in combination with targeted therapy, and the results informed the phase III AMEERA-5 study (ClinicalTrials.gov identifier: NCT04478266), which investigated amcenestrant with palbociclib as first-line therapy in ER+/HER2– aBC.⁴¹ The AMEERA-5 study was discontinued following the results of a futility analysis.

Per FDA guidelines,^42,43 AMEERA-3 included pre-/perimenopausal women with ovarian suppression and men with no previous bilateral orchiectomy pretreated with gonadotropin-releasing hormone. Only one male patient was enrolled and randomly assigned to the TPC arm; therefore, no interpretation of the present trial's results can be made for male patients.

In conclusion, in this open-label, randomized, phase II study in patients with ER+/HER2– aBC, amcenestrant monotherapy administered at a dosage of 400 mg once daily in 28-day cycles was generally well tolerated but did not demonstrate improvement in PFS in the ITT population compared with control patients treated with an AI, fulvestrant, or tamoxifen. However, amcenestrant treatment was associated with a numerically longer PFS in a subset of patients with baseline ESR1 mutation.

DATA SHARING STATEMENT

Qualified researchers may request access to patient-level data and related study documents including the clinical study report, study protocol with any amendments, blank case report form, statistical analysis plan, and data set specifications. Patient-level data will be anonymized, and study documents will be redacted to protect the privacy of our trial participants. Further details on Sanofi's data sharing criteria, eligible studies, and process for requesting access can be found at https://www.vivli.org/.

AUTHOR CONTRIBUTIONS

Conception and design: Sara M. Tolaney, Suzette Delaloge, Hiroji Iwata, Patrick Cohen, Gautier Paux

Provision of study materials or patients: Sara M. Tolaney, Arlene Chan, Suzette Delaloge, Mario Campone, Peter A. Kaufman, Eric Boitier, Seock-Ah Im

Collection and assembly of data: Sara M. Tolaney, Arlene Chan, Katarina Petrakova, Suzette Delaloge, Hiroji Iwata, Peter A. Kaufman, Elisabeth De Kermadec, Patrick Cohen, Gautier Paux, Eric Boitier, Seock-Ah Im

Data analysis and interpretation: Sara M. Tolaney, Arlene Chan, Suzette Delaloge, Mario Campone, Parvin F. Peddi, Peter A. Kaufman, Elisabeth De Kermadec, Qianying Liu, Patrick Cohen, Gautier Paux, Lei Wang, Nils Ternès, Seock-Ah Im

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

AMEERA-3: Randomized Phase II Study of Amcenestrant (Oral Selective Estrogen Receptor Degrader) Versus Standard Endocrine Monotherapy in Estrogen Receptor–Positive, Human Epidermal Growth Factor Receptor 2–Negative Advanced Breast Cancer

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).