Patient-reported outcomes and qualitative interviews in patients with estrogen receptor-positive/human epidermal growth factor receptor 2-negative advanced breast cancer: results from the phase III EMBER-3 trial.

Introduction
Estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative breast cancer is the most common subtype of breast cancer, accounting for 70% of all cases in the United States.1 The mainstay of ER-positive/HER2-negative breast cancer treatment is endocrine therapy (ET),2 which inhibits estrogen production through aromatase inhibitors or ER signaling through selective ER modulators (SERMs) and selective ER degraders (SERDs). In patients with advanced breast cancer (ABC), adding cyclin-dependent kinase (CDK) 4/6 inhibitors to ET has demonstrated substantial progression-free survival (PFS) improvement, thus becoming a preferred treatment option for first-line therapy.2, 3, 4 Most patients, however, eventually develop disease progression. A key mechanism of resistance to ET involves the acquisition of mutations in the estrogen receptor 1 gene (ESR1m), which are found in up to 50% of tumors resistant to aromatase inhibitors and are associated with a shorter PFS compared with ESR1-not detected.5
A common strategy to address endocrine resistance relies on the use of SERDs. Fulvestrant is an intramuscular SERD approved for use, alone or in combination with a CDK4/6 inhibitor, in women with disease progression following aromatase inhibitors.6 In patients with ESR1m, however, fulvestrant exhibits variable efficacy and results in shorter PFS compared with the new oral SERDs.7, 8, 9 The route of administration and limited bioavailability represent important barriers to its use. Consequently, novel SERDs with improved efficacy, ease of oral administration, and favorable safety profiles have been developed for use as monotherapy or as part of combination regimens in ABC.7
Imlunestrant is a next-generation, brain-penetrant, oral SERD that has shown potent antitumor activity against wild-type and mutant ESR1 cells in preclinical studies.10, 11, 12 It binds directly to the ER both wild-type and mutant acting as a pure antagonist, thus blocking estrogen from binding and activating ER signaling including in ESR1-mutant cell lines. Imlunestrant was evaluated alone and with abemaciclib in the phase III, multicenter, open-label randomized, controlled trial (RCT) EMBER-3 in patients with ER-positive/HER2-negative ABC who progressed following ET.13 Compared with standard ET monotherapy (exemestane or fulvestrant) [standard of care (SOC) ET], imlunestrant significantly extended PFS in patients with ESR1m (median PFS: 5.5 versus 3.8 months, P < 0.001). A PFS benefit was also observed with imlunestrant–abemaciclib versus imlunestrant regardless of ESR1m status (median PFS in the concurrently randomized overall population: 9.4 versus 5.5 months, P < 0.001), supporting the use of this oral SERD, alone or with abemaciclib, for the treatment of advanced ER-positive/HER2-negative breast cancer.13 Imlunestrant is now approved by the Food and Drug Administration14 and European Medicines Agency15 for adult patients with ER-positive/HER2-negative, ESR1m advanced or metastatic breast cancer who have progressed after at least one line of ET.
Patients with cancer, especially in the advanced stages, experience disease- and treatment-associated symptoms that can impair their health-related quality of life (HRQL) and functioning. A thorough understanding of patient experiences and perceptions is crucial to inform the overall benefit–risk profile of a treatment,16, 17, 18 particularly in the metastatic setting where the primary goal is to alleviate symptoms and prolong survival while maintaining patient HRQL.2 Patient-reported outcome (PRO) measures are increasingly used in clinical trials to capture patient perspectives on the side-effects and impact of treatment on symptoms, daily functioning, and HRQL.16 Notably, the use of standardized PRO measures to evaluate patient HRQL in clinical trials is recommended by the latest international consensus guidelines for the management of ABC.2
In the EMBER-3 trial, PROs were included as exploratory endpoints to assess the impact of treatment from the patient perspective. In addition, a standalone qualitative interview study was conducted to explore and compare the experiences and perceptions of EMBER-3 trial participants receiving either imlunestrant (oral) or fulvestrant (intramuscular). This approach captures a more nuanced understanding of patient experiences and preferences and is complementary to traditional PRO measures.19,20
Here, we report the findings of exploratory PRO analyses for the EMBER-3 trial. Moreover, we present insights from qualitative patient interviews, highlighting both positive and negative aspects of imlunestrant and fulvestrant administration.

Materials and methods

EMBER-3 trial design and population
The trial design of EMBER-3, together with the efficacy and safety results, have been previously reported.13 Briefly, this was a phase III, multinational, open-label, RCT conducted in women (with any menopausal status) or men aged ≥18 years who had locally confirmed ER-positive/HER2-negative ABC that had recurred or progressed during or after treatment with an aromatase inhibitor (administered alone or with a CDK4/6 inhibitor) as neoadjuvant or adjuvant treatment or while receiving first-line treatment of ABC.13 Patients were initially randomized 1 : 1 to receive oral imlunestrant (400 mg once daily) or SOC ET with oral exemestane (25 mg once daily) or fulvestrant (500 mg, administered as an intramuscular injection on days 1 and 15 of cycle 1 and on day 1 of the subsequent 28-day cycles). Following an amendment early in the enrollment process, patients were randomized 1 : 1 : 1 with the addition of a group that received oral imlunestrant (400 mg once daily) with oral abemaciclib (150 mg twice daily).13 The primary endpoint of EMBER-3 was investigator-assessed PFS with imlunestrant versus SOC ET among the ESR1m population and among all patients, and with imlunestrant–abemaciclib versus imlunestrant among all patients who underwent randomization concurrently.
EMBER-3 was conducted in accordance with the principles of the Declaration of Helsinki, the Good Clinical Practice guidelines of the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use, and applicable laws and regulations.13 The trial protocol was approved by the ethics review board at each site. All participants provided written informed consent.13 The trial is registered with ClinicalTrials.gov (NCT04975308).13

Patient-reported outcomes
In the EMBER-3 trial, PRO measures were included as exploratory endpoints.13 Patients were required to complete the measures on a handheld electronic device at the clinical trial site on day 1 of cycle 1, at home as per the schedule of assessment specified in the protocol, and at site on the last visit.
The European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire Core 30 (QLQ-C30) consists of 30 items with a 7-day recall.21 It comprises a two-item global health status/quality of life (GHS/QoL) scale, five functional scales (15 items) addressing physical, role, social, emotional, and cognitive functioning, and nine symptom scales (13 items) addressing fatigue, pain, nausea/vomiting, dyspnea, insomnia, appetite loss, constipation, and diarrhea. The questionnaire was completed at baseline and every 4 weeks (physical functioning scale) or every 8 weeks (all other subscales). All scales and single items were linearly transformed to 0-100 scales.22 A commonly used definition of clinically meaningful improvement or worsening from baseline in EORTC QLQ-C30 GHS/QoL and functional scores is an increase of ≥10 or a decrease of ≥10 points, respectively, and a clinically meaningful improvement or worsening from baseline in symptom scores is a decrease of ≥10 points or an increase of ≥10 points, respectively.23
The single items assessing diarrhea frequency and injection site reactions (ISR, only for patients receiving fulvestrant) were selected from the PRO version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE) item library.24 Diarrhea frequency was rated on a 5-point scale [0 (never) to 4 (almost constantly)] at baseline and every week. The item characterizing ISR [injection site pain, swelling and redness (yes/no)] was administered to patients treated with fulvestrant every week for cycle 1 and for the first 2 weeks of every cycle thereafter. Recall for both items was the past 7 days.

Qualitative interview study
A global subset of high-enrolling (≥10 patients enrolled) EMBER-3 trial sites in Japan (n = 4), Mexico (n = 2), the United States (n = 1), and Australia (n = 1) identified and screened potential interview participants and obtained their consent to participate. Interviews were conducted from December 2023 to May 2024 in the participant’s native language by a trained interviewer. In qualitative research, there is no defined single method to calculate sample size and power calculations are not appropriate. A best-effort interview sample size of N = 40 was sought to understand the experiences of the target EMBER-3 clinical trial population.25
The interview study was designed and conducted in line with regulatory guidance.20,26 A standalone protocol was developed to conduct semi-structured qualitative interviews among EMBER-3 participants receiving oral imlunestrant monotherapy or intramuscular injections of fulvestrant. A master United States-English protocol was developed for central institutional review board approval in the United States and adapted/translated to meet local requirements. Local, site-specific ethical review board approval of the interview study protocol, separate from clinical trial approval, was obtained. Web-assisted or telephone interviews lasted ∼60 min and included open-ended questions aimed at exploring participants’ feelings before receiving treatment, experiences of receiving imlunestrant or fulvestrant, and the perceived impact of treatment administration mode on daily life. Favorable and unfavorable treatment attributes were explored by asking participants to rank ‘What do you like most about the treatment?’ (1 = best attribute and 3 = worst attribute) and ‘What do you like least about the treatment?’ (1 = worst attribute and 3 = best attribute), respectively, with follow-up questions asked to understand their choices.

Statistical analysis
All PRO analyses were descriptive and no adjustment for multiple testing was carried out. For the EORTC QLQ-C30, the change from baseline in each scale score was calculated through cycles for which ≥25% of patients in each arm had data. A longitudinal mixed model for repeated measures (MMRM) was used to estimate the mean change from baseline [expressed as least squares (LS) mean and 95% confidence interval (CI)] in the EORTC QLQ-C30 score for each treatment arm. The Kaplan–Meier method was used to estimate time to first deterioration (score worsening of ≥10 points) in GHS/QoL. The proportion of patients who worsened (score change of ≥10 points) in all EORTC QLQ-C30 scales was also provided.
For PRO-CTCAE diarrhea frequency, responses were summarized as the number (%) of patients for each category in the questionnaire at each visit, by treatment arm. In addition, the mean (standard deviation, SD) proportion of time during treatment that patients in each treatment arm experienced ‘frequently’ to ‘almost constantly’ diarrhea was estimated. For PRO-CTCAE ISR, the overall proportion of patients that experienced ISR during fulvestrant treatment was provided.
Demographic/health and qualitative interview data were summarized using descriptive statistics and supporting quotes. Interviews were recorded and transcribed directly into English. Data were analyzed using directed content analysis techniques27,28 in ATLAS.ti v9 (ATLAS.ti, Berlin, Germany).

Results

EMBER-3 trial

Patient population
In EMBER-3, between October 2021 and November 2023, 874 patients were randomly assigned to receive imlunestrant monotherapy (n = 331), SOC ET [n = 330, of whom 292 (89%) received fulvestrant] or imlunestrant–abemaciclib (n = 213). Patient characteristics at baseline were generally balanced among treatment arms (Supplementary Table S1, available at https://doi.org/10.1016/j.esmoop.2026.106945).13 Median age was 61 years (range 28-87 years), 62 years (27-89 years), and 62 years (36-87 years) in the imlunestrant, standard ET and imlunestrant–abemaciclib arms, respectively; 195 (59%), 189 (57%), and 139 (65%) patients had previously received a CDK4/6 inhibitor, which was palbociclib in most cases (61%, 69%, and 65%, respectively).13
ESR1 mutations were detected in 138 (42%) patients in the imlunestrant arm and 118 (36%) patients in the standard ET arm.13 The analyses reported here were conducted at the time of the primary analysis (data-cut-off date: 24 June 2024), when the median duration of treatment was 5.6 months with imlunestrant, 4.8 months with standard ET, and 7.7 months with imlunestrant–abemaciclib.13 The main reason for treatment discontinuation was disease progression (72% of patients in the imlunestrant arm, 78% in the standard ET arm, and 53% in the imlunestrant–abemaciclib arm).13

Patient-reported outcomes

EORTC QLQ-C30
Completion rate (patients with an assessable measure among those expected to complete the measure) for the EORTC QLQ-C30 questionnaire was high and similar between treatment arms (86%-88%) at baseline. Completion rates at cycles for which at least 25% of those randomized were still on treatment ranged from 81% to 87% for imlunestrant, 80% to 87% for SOC ET, and 67% to 79% for imlunestrant–abemaciclib. The mean baseline scores for all EORTC QLQ-C30 subscales were generally comparable between treatment arms, with no statistically significant between group differences (Supplementary Table S2, available at https://doi.org/10.1016/j.esmoop.2026.106945).

GHS/QoL, functional, and symptom domains
The EORTC QLQ-C30 GHS/QoL scores were generally maintained (did not exceed 10-point within group change threshold) over time for both treatment arms, with LS mean change values improved for imlunestrant and worsened for SOC ET among the ESR1m population (Figure 1). Time to deterioration of GHS/QoL numerically favored imlunestrant (median 5.6 months, 95% CI 3.8-11.1 months) compared with SOC ET (median 3.8 months, 95% CI 3.0-6.5 months; hazard ratio 0.76, 95% CI 0.5-1.1) (Supplementary Figure S1, available at https://doi.org/10.1016/j.esmoop.2026.106945). The overall LS mean change from baseline within each treatment arm for EORTC QLQ-C30 GHS/QoL, functional, and symptom domains are summarized in Figure 2A-D. GHS/QoL and function were generally maintained across imlunestrant and standard ET among the ESR1m population (Figure 2A) and among all arms in the intention to treat (ITT) population (Figure 2B). Overall, symptoms did not worsen meaningfully for either imlunestrant or SOC ET among the ESR1m population and were broadly similar across treatment arms (Figure 2C). Within the ITT population, symptom domains were broadly similar across treatment arms, except for increased diarrhea and nausea/vomiting with imlunestrant–abemaciclib (Figure 2D). Treatment differences for GHS/QoL, all functional domains, and most symptom domains favored imlunestrant versus SOC among the ESR1m population (Supplementary Figure S2, available at https://doi.org/10.1016/j.esmoop.2026.106945).

Proportion of patients with worsened (≥10 points decrease) EORTC QLQ-C30 GHS/QoL, functional, and symptom domains from baseline
Among patients with ESR1m, a numerically lower proportion of patients in the imlunestrant versus SOC ET arms reported a worsening in EORTC QLQ-C30 GHS/QoL (40% versus 45%), and functional domains from baseline (24%-35% versus 31.3%-46%, respectively), with the exception of the emotional domain (28% versus 24%) (Table 1). In the ITT population, the proportion of patients who reported a worsening in EORTC QLQ-C30 GHS/QoL (42% versus 48%) and functional domains was numerically lower for all domains in the imlunestrant arm (31%-40%) than in the SOC ET arm (36%-45%), except for the emotional domain (29% versus 27%). The proportion of patients treated with imlunestrant–abemaciclib who experienced EORTC QLQ-C30 GHS/QoL and functional domain deterioration (30%-51%) was similar to those observed in the SOC ET arm and marginally higher than in the imlunestrant arm (Table 1).
The proportion of patients with ESR1m who reported symptom worsening was lower or similar in the imlunestrant versus SOC ET arms for the pain, insomnia, appetite loss, and diarrhea domains (24%-42% versus 30%-46%). Worsening in nausea/vomiting, dyspnea, and constipation, however, was more common with imlunestrant than with SOC ET (27%-32% versus 22%-29%) (Table 1). Within the ITT population, symptom deterioration was generally more common in the combination arm (27%-59%) than in the imlunestrant (25%-44%) and SOC ET (26%-47%) arms, especially for appetite loss (45% versus 29% and 31%) and diarrhea (59% versus 28% and 26%) but not for pain (40% versus 40% and 45%) and constipation (27% versus 35% and 27%), respectively (Table 1).

PRO-CTCAE diarrhea frequency
At baseline, most patients in all trial arms reported diarrhea occurring ‘never’, ‘rarely’ or ‘occasionally’; 2.6% of patients in the imlunestrant arm, 1.5% in the standard ET arm, and 3.0% in the imlunestrant–abemaciclib arm reported diarrhea occurring ‘frequently’/‘almost constantly’ (Supplementary Table S2, available at https://doi.org/10.1016/j.esmoop.2026.106945). Throughout the treatment period 140 patients (69%) treated with imlunestrant–abemaciclib reported frequent or almost constant diarrhea at any time, a third of which did so for the majority (≥50% of their PRO-CTCAE assessments) of their time on treatment (Figure 3). Throughout the treatment period, the mean (SD) proportion of assessments for which patients reported experiencing diarrhea ‘frequently’ or ‘almost constantly’ was low and similar between imlunestrant [2.7 (10.1)] and SOC ET [2.3 (7.7)], but higher with imlunestrant–abemaciclib [21.7 (27.6)].

PRO-CTCAE ISR
The incidence of ISR, as measured by the PRO-CTCAE, is shown in Figure 3. Of 278 patients treated with fulvestrant, 201 (72%) reported ISR at any time while on treatment. Of those, 47% reported ISR the majority (≥50% of their PRO-CTCAE assessments) of their time on treatment.

Qualitative interviews study

Patient population
Interview data from 12 eligible EMBER-3 participants (all females) were included in the analysis. Key patient and disease characteristics are reported in Supplementary Table S3, available at https://doi.org/10.1016/j.esmoop.2026.106945. Imlunestrant recipients (n = 8) had a mean (SD) age of 61.3 years (6.2 years), mean length of time since initial breast cancer diagnosis of 10.1 years (6.8 years), and had been enrolled in EMBER-3 for a mean of 1.0 years (0.5 years). Moreover, most patients had high school diploma or equivalent education (n = 6) and were from Japan (n = 3), the United States (n = 2), Mexico (n = 2), and Australia (n = 1). Fulvestrant recipients (n = 4) had a mean (SD) age of 68.3 years (6.2 years), mean length of time since initial breast cancer diagnosis of 10.8 years (4.9 years), and had been enrolled in EMBER-3 for a mean of 2.2 years (1.8 years). Interviewees had a Bachelor’s degree or equivalent (n = 4) and were from Japan (n = 2), the United States (n = 1), and Mexico (n = 1).

Treatment perceptions and experiences
Pre-treatment worries/concerns (e.g. lack of efficacy, potential side-effects) and hope/optimism (for effective treatment) were reported by both imlunestrant and fulvestrant participants. Overall interview participants were positive about the treatment they received during the EMBER-3 trial. This was primarily because, in their opinion, both treatments were efficacious, induced no systemic or only minor side-effects in comparison to other front-line treatments such as chemotherapy, surgery or radiation, and generally had a minimal impact on participants’ daily lives. Subtle differences in experiences in relation to the respective administration method were described (e.g. initial challenges adapting to daily fasting for imlunestrant and minor ISRs for fulvestrant) and participants were open to receiving the alternative monthly injection or daily pill with adequate information or doctor’s advice. Regardless, all participants appeared satisfied with their treatment and were willing and able to accommodate the regimens in their daily lives as adherence to their treatment was important.

Treatment attribute ranking
Participants receiving imlunestrant (daily oral pill) or fulvestrant (monthly intramuscular injection) were asked to complete two ranking tasks at the end of the interview to identify the favorable and unfavorable aspects of the administration mode for each trial medication. Results are summarized in Table 2 and Supplementary Figures S3 and S4, available at https://doi.org/10.1016/j.esmoop.2026.106945.
Among those treated with imlunestrant, eight participants provided favorable attributes and five provided unfavorable attributes. Overall, the most favorable attributes of imlunestrant were the ability to take treatment at home (n = 4) and the simplicity (minimal burden) associated with taking it (n = 4). The most unfavorable attribute was the requirement to fast before and after taking the drug (n = 4).
In the fulvestrant arm, three participants provided favorable and unfavorable attributes. One participant did not complete the fulvestrant attribute task, stating “I don’t feel in terms of whether I like it or dislike it, because it is a therapeutic injection for my cancer.” (07-FU-JP). Overall, the most favorable attributes of fulvestrant administration were the single monthly frequency of the injection (n = 2) and that the injection was administered by qualified staff (n = 2). The most unfavorable attributes were the time burden associated with the need to travel and attend the monthly treatment appointment (n = 2) and pain at the site of the injection (n = 2).

Discussion
PRO measures were included as exploratory endpoints in the EMBER-3 trial to assess the impact of treatment from the patient perspective. In addition, qualitative interviews were conducted to explore and compare the experiences and perceptions of patients receiving oral or intramuscular treatment. Overall, GHS/QoL and function were maintained over time among treatment arms, regardless of ESR1m status. Moreover, fewer patients in the imlunestrant arm reported deterioration of GHS/QoL or function compared with patients in the other treatment arms. Among all symptoms, diarrhea expectedly showed the greatest worsening from baseline with addition of abemaciclib in the imlunestrant–abemaciclib combination arm, and patient-reported diarrhea frequency was consistently higher in the imlunestrant–abemaciclib arm. Most patients treated with fulvestrant reported ISR at some point, with nearly half experiencing it at the majority of assessments. While only a small fraction of those enrolled in the EMBER-3 trial, patients who participated in the interview study considered their experience with imlunestrant or fulvestrant administration positively and, in the context of their metastatic disease, would accept a treatment that worked regardless of administration mode.
Within the ITT population of EMBER-3, GHS/QoL and function were comparable over time between patients on imlunestrant–abemaciclib and those on standard ET. These findings are important when considering the significant prolongation of PFS observed in the combination arm during the trial (median PFS 9.4 months, 95% CI 7.5-11.9 months),13 and support the efficacy of imlunestrant-abemaciclib as an all oral option in patients with ER-positive/HER2-negative ABC.
In the ABC setting, ESR1m is a prognostic marker of worse outcomes.29, 30, 31 In the EMBER-3 trial, patients with ESR1m treated with imlunestrant had generally stable GHS/QoL, functioning, and symptom scores over time, and experienced less frequent GHS/QoL, functioning, and symptom deterioration events than those on standard ET, with a few exceptions. Overall, these results reflect the PFS benefit observed with imlunestrant compared with standard ET in the ESR1m subgroup included in EMBER-3.13
Importantly, 72% of patients treated with fulvestrant reported experiencing ISR at any time, starting from week 1, and 47% experienced ISR at the majority of assessments. This is in contrast with the ISR incidence reported by physicians for patients treated with fulvestrant in EMBER-3 (9%)32 as well as in previous trials (7%-12%).6 The discrepancy between physicians’ and patients’ reports of symptom incidence and severity is well documented and has led to the development of the PRO-CTCAE. The PRO-CTCAE can aid clinical decision-making as it provides a more comprehensive understanding of the tolerability of cancer treatment by complementing physician adverse event (AE) reports, which primarily consider safety (severity, interference, and need for medical intervention), with direct and more reliable reporting of symptoms experienced.33,34
In the EMBER-3 trial, AEs were mostly of grade 1 or 2 in all treatment arms.13 Diarrhea was the most frequent any-grade AE among patients on imlunestrant–abemaciclib, observed in 86% of cases; in 8% of patients, it was of grade ≥3. Therefore, the PRO-CTCAE diarrhea frequency results for patients on imlunestrant–abemaciclib were not unexpected. Further, the increased EORTC QLQ-C30 diarrhea scores observed for patients on imlunestrant–abemaciclib are consistent with those observed with abemaciclib–fulvestrant in MONARCH 2.35
Disease- and treatment-associated symptoms are common in patients with advanced stage disease, and may impair their HRQL, daily activities, and treatment compliance. Symptom burden may be further worsened by the use of combination regimens. Nonetheless, there are symptom management practices for both diarrhea and ISR that may help reduce their impact.36,37
Among the small subset of qualitative interview participants from the EMBER-3 trial, participants generally viewed their experience with imlunestrant and fulvestrant positively, imlunestrant because of being able to take at home and low burden, and fulvestrant because only once monthly and administered by a professional. Regardless of minor challenges related to daily fasting for imlunestrant or injection site pain for fulvestrant, all respondents were willing and able to accommodate the regimens in their daily lives as they acknowledged the importance of treatment adherence. A detailed understanding of patient preferences is important in the decision-making process for the potential impact on treatment adherence and outcomes, particularly in the metastatic setting.38, 39, 40 These findings further support the positive risk–benefit profile of imlunestrant in patients with ABC.
The present analysis has some limitations. Despite good PRO compliance, data availability over time was limited since PRO data were not collected after treatment discontinuation as per study design. This hinders the understanding of patients’ HRQL at and after discontinuation. In addition, this impacts the interpretation of analyses of available data, particularly time to deterioration of function and symptoms, given intercurrent events and the need for serial PRO assessments for long periods of time.41 Therefore, despite the motivation to include time to deterioration analyses in cancer clinical trials because of clinician familiarity with interpreting survival curves and hazard ratios,42 analyses at pre-specified time points using the intended PRO score (continuous or ordinal) may be the preferred approach for evaluating comparative benefit.41
With regard to the qualitative interviews, the final interview sample was just under a third of the initial target due to recruitment challenges, as the interview study began several years into the ongoing clinical trial and was not pre-planned. In qualitative research, however, a sample of 12 participants can still provide valuable insights into participants’ experiences.43 Moreover, the interviews were optional and patients who were well and satisfied may have been more likely to volunteer. Further, for those patients who received imlunestrant, there may have been an expectation bias for a positive experience in knowing that an oral pill would be easier to take than an intramuscular injection. In addition, participants who had discontinued due to progression were excluded, which may result in the perspectives of included participants being more positive than those of other participants. Participants who received imlunestrant–abemaciclib were not included in the qualitative interview study. Understanding the experiences and preferences of patients receiving imlunestrant–abemaciclib, which may differ to those of the interview sample in this study, is a limitation.
Despite these limitations, several strengths should be acknowledged. In EMBER-3, multiple PRO measures were used to provide a comprehensive view of patient experience and HRQL. Moreover, the PRO measures were administered at home via a handheld device and took only a few minutes to complete. The use of devices offers advantages over paper questionnaires, such as reducing patient and administrative burden, improving patient experience and compliance, and minimizing entry errors, potentially leading to more accurate and complete data.44 Lastly, interview participants were recruited from four countries and eight clinical sites, providing valuable perspectives from diverse geographies.

Conclusion
Understanding patient perspectives through PRO measures and qualitative interviews in clinical trials can help inform assessment of the overall benefit–risk profile and actual patient experiences of a treatment. Together, the findings from these exploratory PRO analyses and qualitative patient interview study complement the efficacy and safety data from EMBER-3 and support the use of imlunestrant as monotherapy or combined with abemaciclib, as an all-oral targeted therapy option after progression on ET for patients with ER-positive HER2-negative ABC.