Abstract
Purpose
Adding CDK4/6 inhibitors (CDK4/6is) to endocrine therapy (ET) for HR+/HER2− early breast cancer (EBC) demonstrated statistically significant invasive disease-free survival (iDFS) benefits in monarchE (node positive, high risk, stage II/III) and NATALEE (select N0 and all macroscopic N1, stage II/III). This study evaluated patient preferences for EBC treatment attributes and how these may translate for CDK4/6i selection.
Methods
A web-based discrete choice experiment survey was conducted among US-based adult women with self-reported stage II/III HR+/HER2− EBC. Eight attributes were included, informed by 14 qualitative interviews (to identify most relevant attributes), expert clinical input, and differentiating features between CDK4/6is: efficacy (5-year iDFS), adverse events (venous thromboembolic event [VTE], diarrhea, fatigue), number of blood tests, number of electrocardiograms (EKGs), treatment duration, and schedule. Participants selected scenarios that best reflected their preferences from 10 choice cards, each displaying a pair of hypothetical treatment profiles. A conditional logit regression model was used to estimate preference weights and relative importance (RI) of attributes.
Results
A total of 409 women participated. Patient preferences, from high to low RI, were higher efficacy, lower diarrhea risk, lower fatigue risk, shorter treatment duration, and lower VTE risk. Number of blood tests, number of EKGs, and treatment schedule were less important. Utility scores were higher for reconstructed treatment profiles that resembled ribociclib.
Conclusion
This study demonstrated that patients prefer adjuvant treatment with higher efficacy and lower risk of adverse events. These data will aid shared decision-making when discussing the addition of CDK4/6is to adjuvant ET for eligible patients with HR+/HER2− EBC.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10549-025-07627-4.
Keywords: Patient preference, CDK4/6 inhibitor, Early breast cancer, Discrete choice experiment
Background
Most new breast cancer (BC) cases are diagnosed as early BC (EBC) [1]. Within 20 years of diagnosis, approximately 27–37% of patients with stage II EBC and 46–57% with stage III EBC will experience disease recurrence with standard-of-care (SOC) adjuvant endocrine therapy (ET), which are mostly distant metastatic recurrences [2–4]; half occur within the first 5 years [3, 5]. BC therapies in the adjuvant setting are administered to reduce the risk of recurrence and improve overall survival. Efficacy, safety, and quality of life all contribute to the overall treatment outcomes.
Novel targeted therapies, such as cyclin-dependent kinase 4/6 inhibitors (CDK4/6is), have been studied in the adjuvant setting in patients with hormone receptor-positive, human epidermal growth factor receptor 2-negative (HR+/HER2−) EBC and have shown potential to transform adjuvant care when added to SOC ET. The addition of a CDK4/6i to adjuvant ET for the treatment of HR+/HER2− EBC has been investigated in phase 3 trials of all 3 CDK4/6is that have been approved by the US Food and Drug Administration in the advanced setting [6–9]. In the PALLAS and PENELOPE-B trials, the addition of palbociclib to adjuvant ET did not demonstrate improvement in invasive disease-free survival (iDFS) compared with ET alone [7, 8]. Based on the positive results from monarchE, in 2021 abemaciclib + ET was approved by the US Food and Drug Administration in patients with node-positive HR+/HER2− EBC with a high risk of recurrence [10]. In monarchE, at a median 54-month follow-up, an iDFS benefit with abemaciclib + ET over ET alone was observed in the intention-to-treat population of patients with lymph node-positive, high-risk disease (hazard ratio [HR], 0.68; 95% CI 0.60–0.77) [10]. NATALEE enrolled a broader population of patients with stage II/III HR+/HER2− EBC, including patients with node-negative disease with high-risk features [11]. In an interim analysis of the NATALEE trial, a statistically significant iDFS benefit with ribociclib + a nonsteroidal aromatase inhibitor (NSAI) over NSAI alone (HR, 0.75; 95% CI 0.62–0.91; two-sided P = 0.0003) was observed at a median 27.7-month follow-up [9]. The iDFS benefit with ribociclib + NSAI over NSAI alone was sustained over a median follow-up of 44.2 months (HR, 0.72; 95% CI 0.61–0.84; one-sided P < 0.0001) in an exploratory 4-year landmark analysis [12].
Ribociclib and abemaciclib have been shown to be relatively safe for the treatment of HR+/HER2− BC, with distinct toxicity profiles for each CDK4/6i across the advanced BC (ABC) and EBC settings [13–16]. Higher rates of severe gastrointestinal adverse events (AEs) and fatigue were observed in patients treated with abemaciclib + ET vs ET alone; patients treated with ribociclib reported higher rates of hematologic AEs, neutropenia, elevated liver enzymes, and rare cases of QTc prolongation vs ET alone. These treatment-related AEs necessitated dose modifications and additional monitoring tests for both compounds in clinical trials and clinical practice [13, 14]. In NATALEE, the most common AE associated with ribociclib was neutropenia (any grade, 62.1%; grade ≥ 3, 43.8%); in monarchE, diarrhea was the most common AE associated with abemaciclib (any grade, 83.6%; grade ≥ 3, 7.8%) [9, 17]. These toxicities may lead to premature discontinuation of CDK4/6i. In the NATALEE trial, the most common AEs that led to discontinuation were elevated liver enzymes and arthralgia, and for monarchE, it was diarrhea [9, 17]. With ribociclib, the 400-mg starting dose investigated in the EBC setting was associated with lower rates of neutropenia and QTc prolongation compared with the 600-mg dose, which is the SOC dose in the ABC setting [9, 13]. Abemaciclib and ribociclib also differ in their dosing schedules, which may influence therapy adherence and persistence. In monarchE, patients received either 2 years of abemaciclib (150 mg twice daily on a continuous dosing schedule) + ET (for 5–10 years) or ET alone [6]. In NATALEE, patients received 3 years of ribociclib (400 mg/day; 3 weeks on/1 week off schedule) + ET (for ≥ 5 years) or ET alone [11].
To date, few studies have examined patient preferences for the different attributes of CDK4/6is in BC [18–20]. More research is needed, especially on patient preferences for attributes associated with the 2 CDK4/6is studied in the adjuvant setting, abemaciclib and ribociclib. We report the findings of a discrete choice experiment (DCE) survey that evaluated the extent to which US-based patients with HR+/HER2− EBC value different treatment attributes associated with ribociclib and abemaciclib in the adjuvant setting.
Methods
DCE method overview
The DCE methodology is the most commonly used survey technique to evaluate stated preferences [21–23]. DCE surveys assume that any product or service can be described by a number of attributes (i.e., efficacy, risk of AEs, administration, or dosing) that can be compared and ranked according to patients’ selections [24, 25]. Prior to development of the DCE survey in this analysis, 14 one-on-one qualitative patient interviews were conducted via telephone to elicit the perspective of the US-based patients on their expectations when selecting treatments for adjuvant therapy. Patients also rated the importance they placed on various CDK4/6i treatment attributes on a scale of 1–5, with 5 being most important.
An expert panel consisting of two breast oncologists and a patient advocate was consulted on the overall design and selection of attributes for the DCE survey. Based on the responses from the qualitative interviews and expert panel feedback, a list of eight attributes was considered for the DCE survey: treatment efficacy (5-year iDFS), venous thromboembolic events (VTEs), diarrhea, fatigue, number of blood tests, number of EKGs, treatment duration, and treatment schedule (Table 1). Attributes related to additional AEs were not considered for the DCE based on low ratings in the qualitative interviews; these included the risk of other gastrointestinal AEs, laboratory test result-related AEs such as risk of neutropenia and elevation of liver enzymes, and other rare AEs. The levels reflecting the best and worst values for each attribute were informed by the primary publications of pivotal trials of abemaciclib in the adjuvant EBC setting and ribociclib in the advanced setting, since results from NATALEE were not available at the time of this DCE survey design (Table 1) [10, 15, 26–28].
Table 1.
Treatment attributes and levels
| Features | Treatment efficacy (5-year iDFS) | Venous thromboembolic events | Diarrhea | Fatigue | No. of blood tests | No. of EKGs | Treatment duration | Treatment schedule |
|---|---|---|---|---|---|---|---|---|
| Level 1 | • 90 out of 100 (90%) | • 1 out of 100 (1%) | • 10 out of 100 (10%) | • 10 out of 100 (10%) | • Every 2 weeks for the first 2 months, monthly for the next 2 months | • No EKGs required | • 2 years | • Once daily, intermittent treatment with a 1-week break each month |
| Level 2 | • 85 out of 100 (85%) | • 3 out of 100 (3%) | • 75 out of 100 (75%) | • 40 out of 100 (40%) | • Every 2 weeks for the first 2 months, monthly for the next 4 months | • 3 EKGs in the first month | • 3 years | • Twice daily, continuous treatment |
| Level 3 | • 80 out of 100 (80%) | |||||||
| Level 4 | • 75 out of 100 (75%) |
EKG electrocardiogram
Survey methods
Women aged ≥ 18 years diagnosed with HR+/HER2− stage II/III EBC, living in the USA, and receiving adjuvant ET at the time of completing the survey were enrolled. Participants who received prior (neo)adjuvant chemotherapy were eligible but were required to have stopped chemotherapy at the time of enrollment. Participants’ most recent BC surgery was required to be within 1–3 years prior to the enrollment date. Participants were excluded if they had been diagnosed with metastatic BC (stage IV); were first diagnosed with BC ≥ 5 years ago; had previously received a CDK4/6i; were currently enrolled in a BC clinical trial; or had previously received or were currently receiving treatments indicated for ABC only or other subtypes of BC (i.e., triple negative or HER2 +), including mTOR inhibitors, PD-L1 inhibitors, monoclonal antibodies/antibody–drug conjugates, or kinase inhibitors.
A web-based DCE survey was conducted among participants between January and May 2023, before NATALEE results were available. Email invitations were sent to panel members and potential participants via physician networks, patient associations, and social media outreach, including a URL to a secure website link that hosted the survey and included a series of pre-survey questions to determine eligibility. Informed consent was collected from participants at the time of enrollment. Patients were incentivized to complete the survey. In addition to answering the DCE survey questions, participants provided self-reported information on demographics, socioeconomic characteristics, and disease history. Prior to the survey, participants were provided with a tutorial to help them understand the attributes and choice card design using patient-friendly language that was reviewed by the patient advocate. Prompts were implemented to ensure that participants did not rush to complete the survey, and data quality checks were conducted to exclude these participants. The DCE questions were presented as choice cards that consisted of pairs of hypothetical treatment profiles with varying levels for each attribute, and participants were asked to select the treatment option that best reflected their preference (Supplemental Fig. 1). In total, 32 choice cards were generated based on efficient design by optimizing the D-efficiency in SAS (Statistical Analysis System Enterprise Guide 7.1). To minimize responder bias, cards were randomized into four blocks, with two additional cards for validity checks (one for dominance test and one for stability test) in each block. For the dominance test, attribute levels in one treatment profile (dominant profile) were objectively better than those in the other profile (dominated profile); participants were expected to choose the dominant profile. For the stability test, participants were asked to respond to two choice cards with the same treatment profiles in reverse order; participants were expected to choose the same treatment profile on both cards. Participants who failed the dominance and stability tests were excluded in the sensitivity analysis. Only de-identified data were collected. Institutional review board exemption was received prior to the qualitative interviews and survey data collection.
Statistical analysis
A conditional logit regression model was estimated to analyze patients’ preferences between alternative treatment profiles and their willingness to trade off between these attributes. The dependent variable for the model was the patient’s preferred choice, and the independent variables were the attributes with pre-specified levels of the choice cards. Coefficients obtained from regression analyses were used to calculate the relative importance (RI) of the attributes and associated P values, indicating whether a specific attribute has any significant impact in terms of decision-making. RI scores were estimated as normalized percentages calculated by multiplying the coefficient for each attribute by the difference between the best and worst levels of the attribute, to allow for comparisons across attributes. The RI score for an attribute could be interpreted as the proportional value of a change from a treatment with the worst level to a treatment with the best level. A minimum acceptable benefit was calculated to estimate the minimum change in 5-year iDFS benefit required by patients to accept a treatment with a less desirable level of risk (best to worst level). An overall utility score (estimated overall preference) was derived from the model for treatment profiles that resembled abemaciclib and ribociclib profiles (assuming equal efficacy in the base case), including one conservative efficacy scenario: a 10% lower iDFS rate than in abemaciclib-like profiles to represent profiles resembling ribociclib (Supplemental Table 1) [10, 26]. Subgroup analyses were conducted for pre-/perimenopausal and postmenopausal status, as well as stage II BC and stage III BC. A sensitivity analysis was conducted by excluding patients who failed the stability and/or dominance tests to ensure robustness of results [29].
Results
Patient characteristics
Data for the DCE were collected from 409 US-based women with HR+/HER2− EBC treated in the adjuvant setting; 49% of participants were pre-/perimenopausal and 48% had stage II disease at the time of the survey (Table 2). The mean (SD) age in the overall sample was 54.5 (7.8) years, and the median age was 53.0 years. In the overall sample, 23% of participants were Black or African American, and approximately half of the participants were White (59%). The majority of participants were from the western region of the USA (40%), followed by the southern (25%), northeastern (15%), and mid-western (8%) regions. A quarter of the survey participants were employed either full time or part time (38%).
Table 2.
Patient characteristics
| Characteristics | Overall N = 409 |
Subgroups by stage | |
|---|---|---|---|
| Stage II n = 197 |
Stage III n = 212 |
||
| Age, mean ± SD [median], years | 54.5 ± 7.8 [53.0] | 52.6 ± 7.9 [51.0] | 56.2 ± 7.3 [56.0] |
| Race, n (%)a | |||
| White | 240 (58.7) | 122 (61.9) | 118 (55.7) |
| Black or African American | 94 (23.0) | 41 (20.8) | 53 (25.0) |
| Other/prefer not to answer | 92 (22.5) | 47 (23.9) | 45 (21.2) |
| Time since first breast cancer diagnosis, n (%)b | |||
| < 2 years ago | 112 (27.4) | 76 (38.6) | 36 (17.0) |
| 2 to < 3 years ago | 165 (40.3) | 72 (36.5) | 93 (43.9) |
| 3 to < 5 years ago | 126 (30.8) | 49 (24.9) | 77 (36.3) |
| Menopausal status, n (%)c | |||
| Premenopausal or perimenopausal | 199 (48.7) | 126 (64.0) | 73 (34.4) |
| Postmenopausal | 209 (51.1) | 71 (36.0) | 138 (65.1) |
| BC stage at survey, n (%)c | |||
| II | 197 (48.2) | 197 (100.0) | 0 |
| III | 212 (51.8) | 0 | 212 (100.0) |
| Type of prior treatment received, n (%)a | |||
| Chemotherapy | 286 (69.9) | 136 (69.0) | 150 (70.8) |
| Endocrine/hormonal treatmentd | 235 (57.5) | 112 (56.9) | 123 (58.0) |
| PARP inhibitors | 24 (5.9) | 6 (3.0) | 18 (8.5) |
| Not sure | 12 (2.9) | 0 | 12 (5.7) |
| Employment status, n (%)b | |||
| Employed | 154 (37.7) | 77 (39.1) | 80 (37.7) |
| Unemployed | 246 (60.1) | 120 (60.9) | 126 (59.4) |
| US census region, n (%) | |||
| Northeast | 61 (14.9) | 25 (12.7) | 36 (17.0) |
| Mid-west | 34 (8.3) | 13 (6.6) | 21 (9.9) |
| South | 104 (25.4) | 43 (21.8) | 61 (28.8) |
| West | 163 (39.9) | 91 (46.2) | 72 (34.0) |
| Prefer not to answer | 47 (11.5) | 25 (12.7) | 22 (10.4) |
BC breast cancer, PARP poly (ADP-ribose) polymerase, SD standard deviation
aCategories are not mutually exclusive
bResponses of “Not sure” and “Prefer not to answer” are not shown
cAt survey completion
d100% of patients selected for the DCE were on an ET at the time of survey; some of these patients reported receiving an endocrine/hormonal therapy in the past
Patient preferences in overall sample
In the overall sample of 409 patients, in decreasing order of RI, higher efficacy (iDFS), lower risk of diarrhea, lower risk of fatigue, shorter treatment duration, and lower risk of VTE had a significant influence on participant’s treatment preference (Fig. 1, Table 3). When making treatment decisions, participants placed the highest importance on treatment efficacy (RI: 33%; P < 0.001). An increased risk of AEs negatively impacted patients’ preferences. Among the AE attributes, risk of diarrhea was most important (RI: 26%; P < 0.001), followed by fatigue (RI: 14%; P < 0.001) and VTE (RI: 8%; P < 0.001). Shorter treatment duration positively influenced participants’ treatment preference but was of lower RI than risk of diarrhea or fatigue (RI: 12%; P < 0.001). Treatment schedule (RI: 0.9%; P = 0.519), number of blood tests (RI: 3%; P = 0.071), and number of EKGs (RI: 2%; P = 0.146) did not significantly influence participants’ treatment preference. On average, participants would require at least a 3.7-percentage-point increase in 5-year iDFS to tolerate a 2.0-percentage-point increase in risk of VTE (the difference between the worst and best values for the attribute) (Table 4). Additionally, to tolerate a 61.8-percentage-point increase in risk of diarrhea or an 18.6-percentage-point increase in risk of fatigue, patients would require at least an 11.2- or 4.0-percentage-point increase in 5-year iDFS, respectively.
Fig. 1.
RI ranking of treatment attributes in overall samplea. EKG electrocardiogram, iDFS invasive disease-free survival, RI relative importance. aThis radar chart demonstrates the RI of each attribute, with 1 being the highest RI and 8 the lowest
Table 3.
Regression output (preference weights), willingness to trade off, and RI in overall sample
| Attribute | Coefficient (95% CI) | P value | Willingness to trade offa | RI |
|---|---|---|---|---|
| Treatment efficacy (iDFS) | 8.336 (6.98, 9.70) | < .001b | – | 1.25 (33%) |
| AEs | ||||
| Venous thromboembolic events | − 15.455 (− 21.02, − 9.89) | < .001b | 9.27 | 0.31 (8%) |
| Diarrhea | − 1.504 (− 1.72, − 1.29) | < .001b | 0.90 | 0.98 (26%) |
| Fatigue | − 1.773 (− 2.16, − 1.39) | < .001b | 1.06 | 0.53 (14%) |
| Number of blood tests | ||||
| Every 2 wk for the first 2 mo, monthly for the next 2 mo (Ref) | – | – | – | – |
| Every 2 wk for the first 2 mo, monthly for the next 4 mo | 0.116 (− 0.01, 0.24) | .071 | − 0.01 | 0.12 (3%) |
| Number of EKGs | ||||
| None required (Ref) | – | – | – | – |
| 3 in the first month | 0.088 (− 0.03, 0.21) | .146 | − 0.01 | 0.09 (2%) |
| Treatment duration | ||||
| 2 years (Ref) | – | – | – | – |
| 3 years | − 0.461 (− 0.60, − 0.33) | < .001b | 0.06 | 0.46 (12%) |
| Treatment schedule | ||||
| Once daily, intermittent treatment with a 1-wk break each mo (Ref) | – | – | – | – |
| Twice daily, continuous treatment | − 0.034 (− 0.14, 0.07) | .519 | 0 | 0.03 (0.9%) |
AE adverse event, CI confidence interval, EKG electrocardiogram, iDFS invasive disease-free survival, RI relative importance
aWillingness to trade off was calculated using the formula − (βattribute/βiDFS). The number indicates how many percentage points of iDFS at a certain time point an average patient is willing to trade off to avoid a particular attribute. For AEs, the trade-off is expressed with respect to a 5-percentage-point reduction in the risk of having the AE; that is, willingness to trade off was calculated using the formula − (βattribute/βiDFS) × 5
bStatistically significant at the 5% level
Table 4.
Minimum acceptable benefit in overall sample
| Attribute | Minimum acceptable 5-year iDFS benefit, % |
|---|---|
| Venous thromboembolic events | |
| Increase from 1.5 to 2.5% | 1.9 |
| Increase from 1.0 to 3.0% | 3.7 |
| Diarrhea | |
| Increase from 13.9 to 75.7% | 11.2 |
| Increase from 10.0 to 75.0% | 11.7 |
| Fatigue | |
| Increase from 22.3 to 40.9% | 4.0 |
| Increase from 10.0 to 40.0% | 6.4 |
| Treatment duration | |
| Change from 2 to 3 years | 5.5 |
iDFS invasive disease-free survival
Patient preferences in subgroups
Subgroup results by BC stage and menopausal status were generally consistent with results in the overall sample (Fig. 2, Supplemental Tables 2 and 3). Participants with stage III BC placed higher RI on fatigue vs treatment duration (similar to the relative ordering of these attributes in the overall sample), while the relative ordering was reversed among participants with stage II BC. These findings may be due to a higher proportion of pre-/perimenopausal patients with stage II BC (64%) among the stage II subgroup and a higher proportion of postmenopausal patients with stage III BC (65%).
Fig. 2.
RI ranking of treatment attributes in breast cancer stage subgroupsa. EKG electrocardiogram, RI relative importance. aThis radar chart demonstrates the RI of each attribute, with 1 being the highest RI and 8 the lowest
Utility analysis
Overall, treatment profiles in the DCE survey that resembled the profile of ribociclib had consistently higher utility scores, despite the higher preference for a lower treatment duration, including a scenario in which treatment efficacy was assumed to be 10% lower with profiles resembling ribociclib (Table 5). The differences in utility scores were primarily driven by differences in the risk of diarrhea and fatigue. Similar results for utility differences were also seen in subgroups by BC stage or menopausal status, showing a stronger preference for profiles resembling ribociclib in patients with EBC (Supplemental Tables 4 and 5).
Table 5.
Utility reconstruction for profiles resembling ribociclib or abemacicliba
| Attribute | Utility for ribociclib-like profile | Utility for abemaciclib-like profile | Difference in utility |
|---|---|---|---|
| Treatment efficacy (5-year iDFS) | |||
| Base case | 6.97 | 6.97 | 0 |
| Scenario 1 | 6.14 | 6.97 | − 0.83 |
| AEs | |||
| Venous thromboembolic events | − 0.23 | − 0.39 | 0.15 |
| Diarrhea | − 0.21 | − 1.14 | 0.93 |
| Fatigue | − 0.40 | − 0.73 | 0.33 |
| Number of blood tests | |||
| Every 2 wk for the first 2 mo, monthly for the next 2 mo (Ref) | – | – | – |
| Every 2 wk for the first 2 mo, monthly for the next 4 mo | 0.12 | 0 | 0.12 |
| Number of EKGs | |||
| None required (Ref) | – | – | – |
| 3 in the first month | 0.09 | 0 | 0.09 |
| Treatment duration | |||
| 2 years (Ref) | – | – | – |
| 3 years | − 0.46 | 0 | − 0.46 |
| Treatment schedule | |||
| Once daily, intermittent treatment with a 1-wk break each mo (Ref) | – | – | – |
| Twice daily, continuous treatment | 0 | − 0.03 | 0.03 |
| Profile utility | |||
| Base case | 5.87 | 4.69 | 1.19 |
| Scenario 1 | 5.04 | 4.69 | 0.36 |
AE adverse events, EKG electrocardiogram, iDFS invasive disease-free survival
aRegression coefficients from a multivariable logistic model that regresses patients’ preference choice on the attribute’s level on the choice cards
Sensitivity analysis
Results from sensitivity analyses, when excluding participants who failed dominance and/or stability tests, remained consistent with the main analyses, indicating the robustness of the findings (data not shown).
Discussion
This analysis used a DCE survey to assess patient preferences for attributes related to the efficacy, safety, and dose administration of CDK4/6is in the adjuvant setting among patients with HR+/HER2− EBC. In order of importance, higher efficacy (assessed by iDFS rate at 5 years), lower risk of diarrhea, lower risk of fatigue, shorter treatment duration, and lower risk of VTE had a significant influence on participant treatment preferences. Attributes related to the number of blood tests needed to monitor laboratory abnormalities and EKGs to monitor signs of abnormal heart rhythm did not significantly influence patients’ preferences for treatments. Overall, among patients with EBC, treatment profiles that approximated the profile of ribociclib had a higher estimated average utility, despite patients’ preference for shorter treatment duration. The same preferences were generally evident across subgroups by menopausal status or BC stage. These findings emphasize the importance of considering the preferences of patients who are eligible for monarchE (node positive, high risk, stage II/III) or NATALEE (select N0 and all macroscopic N1, stage II/III) along with other relevant factors (e.g., patient history, drug–drug interactions, comorbid clinical conditions, and clinical experience managing AEs) when clinicians and patients are making treatment decisions together.
A previously published DCE survey of attributes associated with CDK4/6is + ET vs ET alone, including oncologists and patients with stage II/III HR+/HER2− EBC who had received prior ET in the adjuvant setting, identified iDFS as the most important attribute to patients and oncologists [18]. In that analysis, to accept an increased risk of diarrhea (from 11% [lowest attribute level] to 81% [highest attribute level]), oncologists and patients required an improvement in iDFS of 5.6 and 8.0 percentage points over the base level, respectively; patients were less willing to trade off a higher risk of diarrhea for a more modest improvement in efficacy. Additional studies in the advanced setting have also reported consistent findings on patients’ preference for a lower risk of symptomatic AEs [19]. A survey of healthcare providers and patients with HR+/HER2− ABC found that many patients were unwilling to tolerate AEs (e.g., fatigue, pain, diarrhea, and loss of appetite), even if their treatment was effective [30].
Some limitations of this DCE survey need to be acknowledged. Since the DCE was designed prior to availability of the NATALEE trial results, the levels of the safety attributes were informed by metastatic BC trials of ribociclib in addition to the safety results from monarchE. The use of safety data from metastatic trials may be perceived as a limitation; however, the wide range for the attribute levels to ensure conservativeness could have mitigated any impact of this on the results. To reduce responder burden, the DCE survey did not consider all possible attributes. Some, such as risk of neutropenia or liver enzyme elevation, could be important to clinicians, and their exclusion may be perceived as a limitation. These attributes were excluded from the DCE survey based on their low ranking (2.4 out of 5 for neutropenia; 2.8 out of 5 for liver enzyme elevation) by participants in the 14 qualitative interviews conducted prior to the DCE survey. To potentially mitigate this limitation, the AEs and the potential for requiring further dose adjustments or interruptions were explained to participants in the DCE survey tutorial in the context of the number of blood tests required for monitoring these AEs. However, this did not impact the RI that participants assigned to blood test monitoring. Patient out-of-pocket (OOP) costs were not considered as an attribute in this DCE survey, which could be perceived as a limitation given the high sensitivity to drug prices. In the real world, the OOP cost burden can vary significantly based on several factors, including drug prices, the number of medicines a patient is taking, health insurance coverage, and accessibility of OOP cost assistance programs. A high OOP cost burden could impact patients’ preference for treatments. In general, patients who participate in online survey studies tend to be more educated, to be younger, and to have better health status than the average patient with EBC. Furthermore, the findings of this study are only applicable to patients who are considering adding a CDK4/6i to adjuvant ET; in the real world, some patients will not consider an add-on treatment to ET. Hence, the overall sample in this DCE survey may not be fully representative of the larger stage II/III EBC population in the USA. The overall sample size enrolled in this DCE survey is sufficiently robust to mitigate any significant bias due to a less representative sample. Lastly, DCE surveys inherently rely on the assumption that participants make rational choices, whereas this may not always be true in the real world. Sensitivity analyses were conducted to ensure internal validity of the results.
This DCE survey showed that patients with EBC placed greater importance on efficacy and lower risk of symptomatic AEs (diarrhea, fatigue, and VTE) over other included attributes when making treatment choices based on attributes of CDK4/6is in the adjuvant setting. These findings suggest that patients with stage II/III HR+/HER2− EBC may exhibit a greater preference for treatment profiles that resemble ribociclib. These findings could aid in physician–patient combined decision-making when discussing the addition of a CDK4/6i to SOC adjuvant treatment for eligible patients with HR+/HER2− EBC.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We thank the participants of this study. We thank Daniele Cary, PhD, of Nucleus Global, for medical editorial assistance on this manuscript.
Author contributions
The study concept and design were developed by A.G., D.L.-V., N.H., V.G., and L.S. Acquisition, analysis, and interpretation of data were performed by A.G., D.L.-V., E.M., M.L., N.H., Y.M., W.Q., V.G., and L.S. Statistical analysis was performed by A.G., D.L.-V., N.H., W.Q., and R.B. Study supervision was performed by V.G. The main protocol authors were E.M., A.G., D.L.-V., N.H., V.G., W.Q., and M.L. All authors read and approved the final manuscript. All authors commented on previous versions of the manuscript, and all authors read and approved the final manuscript.
Funding
This work was supported by Novartis, which also provided financial support for medical editorial assistance.
Data availability
The data that support the findings of this study are not publicly available because the participants did not consent to their data being analyzed outside of this study.
Declarations
Competing interest
Erica Mayer received consulting fees from Eli Lilly, AstraZeneca, Novartis, and Arvinas. All other authors have no relevant financial or non-financial interests to disclose.
Ethical approval
The study was conducted in accordance with ISPOR Good Research Practices for Conjoint Analysis. The DCE survey did not collect any personal identifiable information; institutional review board exemption was received prior to data collection.
Consent to participate
Informed consent was collected from all participants at the time of enrollment.
Consent to publish
The authors affirm that human research participants provided informed consent for publication of all data.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Iqbal J, Ginsburg O, Rochon PA, Sun P, Narod SA (2015) Differences in breast cancer stage at diagnosis and cancer-specific survival by race and ethnicity in the United States. JAMA 313(2):165–173 [DOI] [PubMed] [Google Scholar]
- 2.Pan H, Gray R, Braybrooke J, Davies C, Taylor C, McGale P et al (2017) 20-year risks of breast-cancer recurrence after stopping endocrine therapy at 5 years. N Engl J Med 377(19):1836–1846 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gomis RR, Gawrzak S (2017) Tumor cell dormancy. Mol Oncol 11(1):62–78 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Early Breast Cancer Trialists' Collaborative Group (EBCTCG) (2022) Aromatase inhibitors versus tamoxifen in premenopausal women with oestrogen receptor-positive early-stage breast cancer treated with ovarian suppression: a patient-level meta-analysis of 7030 women from four randomised trials. Lancet Oncol 23(3):382–392 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Foldi J, O’Meara T, Marczyk M, Sanft T, Silber A, Pusztai L (2019) Defining risk of late recurrence in early-stage estrogen receptor-positive breast cancer: clinical versus molecular tools. J Clin Oncol 37(16):1365–1369 [DOI] [PubMed] [Google Scholar]
- 6.Johnston SRD, Harbeck N, Hegg R, Toi M, Martin M, Shao ZM et al (2020) Abemaciclib combined with endocrine therapy for the adjuvant treatment of HR+, HER2-, node-positive, high-risk, early breast cancer (monarchE). J Clin Oncol 38(34):3987–3998 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Mayer EL, Dueck AC, Martin M, Rubovszky G, Burstein HJ, Bellet-Ezquerra M et al (2021) Palbociclib with adjuvant endocrine therapy in early breast cancer (PALLAS): interim analysis of a multicentre, open-label, randomised, phase 3 study. Lancet Oncol 22(2):212–222 [DOI] [PubMed] [Google Scholar]
- 8.Loibl S, Marmé F, Martin M, Untch M, Bonnefoi H, Kim SB et al (2021) Palbociclib for residual high-risk invasive HR-positive and HER2-negative early breast cancer—the Penelope-B trial. J Clin Oncol 39(14):1518–1530 [DOI] [PubMed] [Google Scholar]
- 9.Slamon D, Lipatov O, Nowecki Z, McAndrew N, Kukielka-Budny B, Stroyakovskiy D et al (2024) Ribociclib plus endocrine therapy in early breast cancer. N Engl J Med 390(12):1080–1091 [DOI] [PubMed] [Google Scholar]
- 10.Rastogi P, O’Shaughnessy J, Martin M, Boyle F, Cortes J, Rugo HS et al (2024) Adjuvant abemaciclib plus endocrine therapy for hormone receptor–positive, human epidermal growth factor receptor 2–negative, high-risk early breast cancer: results from a preplanned monarchE overall survival interim analysis, including 5-year efficacy outcomes. J Clin Oncol 42(9):987–993 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Slamon DJ, Fasching PA, Hurvitz S, Chia S, Crown J, Martín M et al (2023) Rationale and trial design of NATALEE: a phase III trial of adjuvant ribociclib + endocrine therapy versus endocrine therapy alone in patients with HR+/HER2- early breast cancer. Ther Adv Med Oncol 15:17588359231178124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Fasching PA, Stroyakovskiy D, Yardley D, Huang CS, Crown JP, Bardia A et al (2024) LBA13 adjuvant ribociclib (RIB) plus nonsteroidal aromatase inhibitor (NSAI) in patients (Pts) with HR+/HER2− early breast cancer (EBC): 4-year outcomes from the NATALEE trial. Ann Oncol 35:S1207 [Google Scholar]
- 13.Burris HA, Chan A, Bardia A, Thaddeus Beck J, Sohn J, Neven P et al (2021) Safety and impact of dose reductions on efficacy in the randomised MONALEESA-2, -3 and -7 trials in hormone receptor-positive, HER2-negative advanced breast cancer. Br J Cancer 125(5):679–686 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Rugo HS, Huober J, García-Sáenz JA, Masuda N, Sohn JH, Andre VAM et al (2021) Management of abemaciclib-associated adverse events in patients with hormone receptor-positive, human epidermal growth factor receptor 2-negative advanced breast cancer: safety analysis of MONARCH 2 and MONARCH 3. Oncologist 26(1):e53–e65 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Barrios CH, Harbeck N, Hortobagyi GN, O’Shaughnessy J, Huang CS, Jimenez MM et al (2024) 113MO NATALEE update: safety and treatment (tx) duration of ribociclib (RIB) + nonsteroidal aromatase inhibitor (NSAI) in patients (pts) with HR+/HER2− early breast cancer (EBC). ESMO Open 9:103101 [Google Scholar]
- 16.Rugo HS, O’Shaughnessy J, Boyle F, Toi M, Broom R, Blancas I et al (2022) Adjuvant abemaciclib combined with endocrine therapy for high-risk early breast cancer: safety and patient-reported outcomes from the monarchE study. Ann Oncol 33(6):616–627 [DOI] [PubMed] [Google Scholar]
- 17.Johnston SRD, Toi M, O’Shaughnessy J, Rastogi P, Campone M, Neven P et al (2023) Abemaciclib plus endocrine therapy for hormone receptor-positive, HER2-negative, node-positive, high-risk early breast cancer (monarchE): results from a preplanned interim analysis of a randomised, open-label, phase 3 trial. Lancet Oncol 24(1):77–90 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Beusterien K, Maculaitis MC, Hallissey B, Gaschler MM, Smith ML, Law EH (2021) Patient, oncologist, and payer preferences for adjuvant endocrine therapy and CDK4/6 inhibitor regimens in early-stage breast cancer: a discrete choice experiment. Patient Prefer Adherence 15:611–623 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Omori Y, Enatsu S, Cai Z, Ishiguro H (2019) Patients’ preferences for postmenopausal hormone receptor-positive, human epidermal growth factor receptor 2-negative advanced breast cancer treatments in Japan. Breast Cancer 26(5):652–662 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Maculaitis MC, Liu X, Will O, Hanson M, McRoy L, Berk A, Crastnopol M (2020) Oncologist and patient preferences for attributes of CDK4/6 inhibitor regimens for the treatment of advanced/metastatic HR positive/HER2 negative breast cancer: discrete choice experiment and best-worst scaling. Patient Prefer Adherence 14:2201–2214 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Soekhai V, de Bekker-Grob EW, Ellis AR, Vass CM (2019) Discrete choice experiments in health economics: past, present and future. Pharmacoeconomics 37(2):201–226 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bridges JF, Hauber AB, Marshall D, Lloyd A, Prosser LA, Regier DA et al (2011) Conjoint analysis applications in health–a checklist: a report of the ISPOR Good Research Practices for Conjoint Analysis Task Force. Value Health 14(4):403–413 [DOI] [PubMed] [Google Scholar]
- 23.Reed Johnson F, Lancsar E, Marshall D, Kilambi V, Mühlbacher A, Regier DA et al (2013) Constructing experimental designs for discrete-choice experiments: report of the ISPOR Conjoint Analysis Experimental Design Good Research Practices Task Force. Value Health 16(1):3–13 [DOI] [PubMed] [Google Scholar]
- 24.Marshall D, Bridges JF, Hauber B, Cameron R, Donnalley L, Fyie K, Johnson FR (2010) Conjoint analysis applications in health—how are studies being designed and reported? An update on current practice in the published literature between 2005 and 2008. Patient 3(4):249–256 [DOI] [PubMed] [Google Scholar]
- 25.Ryan M, Gerard K (2003) Using discrete choice experiments to value health care programmes: current practice and future research reflections. Appl Health Econ Health Policy 2(1):55–64 [PubMed] [Google Scholar]
- 26.Hortobagyi G, Stroyakovsky D, Yardley D, Huang C-S, Fasching PA, Crown J et al (2024) Abstract GS03-03: Ribociclib (RIB) + nonsteroidal aromatase inhibitor (NSAI) as adjuvant treatment in patients with HR+/HER2− early breast cancer: final invasive disease–free survival (iDFS) analysis from the NATALEE trial. Cancer Res 84:GS03-03 [Google Scholar]
- 27.Kisqali® [Prescribing Information] (2022) East Hanover: Novartis Pharmaceuticals Corporation. https://www.novartis.us/sites/www.novartis.us/files/kisqali.pdf
- 28.Verzenio® [Prescribing Information] (2023) Indianapolis: Eli Lilly and Company
- 29.Ryan M (2004) Discrete choice experiments in health care. BMJ 328(7436):360–361 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Cardoso F, Rihani J, Aubel D, de Courcy J, Harmer V, Harbeck N et al (2022) 178P Assessment of side effects (SEs) impacting quality of life (QoL) in patients (pts) undergoing treatment (tx) for advanced breast cancer (ABC) in clinical practice: a real-world (RW) multi-country survey. Ann Oncol 33:S208–S209 [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are not publicly available because the participants did not consent to their data being analyzed outside of this study.