Disease landscape of advanced HER2-breast cancer patients by treatment line in three EU countries and USA

Abstract

Aim: To report treatment patterns and quality of life (QoL) in HER2-negative advanced breast cancer patients. Methods: Data were drawn from a cross-sectional survey in Europe and USA. Results: Hormone plus targeted therapy was the most frequent first-line (1L, 62%) and second-line (2L, 45%) treatment for HR+/HER2-patients. Chemotherapy was most frequent at third-line or greater (3L+, 39%) for HR+/HER2- patients, 2L (51%) and 3L+ (48%) for triple negative breast cancer (TNBC) patients. Time to progression was 13.8 (2L) and 11.0 (3L+) months for HR+/HER2- patients. No comparisons were observed for TNBC patients. EQ-5D-5L scores were highest in patients at 1L and lowest at 3L+. Conclusion: Reduced QoL and treatment response were reported in patients at later lines of therapy.

Plain language summary

Breast cancer is the most common cancer in women. Differences in survival are seen depending on how widespread or advanced the cancer is, how many different treatments the patient has been given, as well as whether certain receptors on the tumor are present or absent. Many new treatments are available which can target these receptors. These treatments have improved survival in patients with advanced breast cancer, but other benefits for the patient are not always clear. In addition, differences between countries are possible as official guidance can vary. This study aimed to understand these issues, by asking physicians and their patients across Europe and USA for their views on quality of life and satisfaction with their treatments. We found that, in general, physicians prescribed treatments as recommended in the treatment guidelines. As breast cancer progressed and treatment stopped working, patients were switched on to different treatments. Survival, quality of life and treatment satisfaction were all worse in patients who had switched treatments. It appears that the patients lose confidence that their new treatment will work to improve their quality of life. We also saw differences in some of these outcomes between Europe and USA, which were likely due to differences in the treatment guidelines between countries. Both quality of life and treatment satisfaction are important for the well-being of patients with advanced breast cancer as they now live longer with these new treatments. This should be considered by physicians and taken into account for future work.

Summary points

• Given the rapidly evolving treatment landscape for patients with advanced breast cancer (ABC), the aim of this analysis was to describe treatment patterns, quality of life (QoL) outcomes, and treatment satisfaction at subsequent treatment lines (first- [1L], second- [2L]; third/later- [3L+] lines) among patients with hormone-receptor positive (HR+)/human epidermal growth factor-negative (HER2-) and triple negative breast cancer (TNBC) in Europe (France, Italy and Spain) and USA in a real-world setting.

• Data were drawn from the Advanced Breast Cancer (ABC) Disease Specific Programme^™, a point-in-time survey of physicians and their consulting patients conducted in Europe and the USA 2019–2020.

• Median time to progression at 2L and 3L was 13.8 and 11.0 months for patients with HR+/HER2, and 8.0 and 4.6 months, respectively for TNBC patients; patient overall response rates at 1L, 2L and 3L+ were 54, 53 and 42%, respectively.

• The average EuroQol-5 dimensions-5 level (EQ-5D-5L) health utility score of HR+/HER2- and TNBC patients was 0.77 at 1L, 0.70 and 0.71 at 2L, and 0.65 and 0.79 at 3L+, respectively; EQ-5D-5L visual analog scale mean score was 70.9, 67.4, and 63.4 for HR+/HER2- patients, and 69.4, 65.2, and 65.8 for TNBC patients, at 1L, 2L and 3L+, respectively.

• The Quality of Life Questionnaire-Core 30 (EORTC QLQ-C30) global health status mean score was 64.6, 60.0 and 57.7 for HR+/HER2- patients, and 64.0, 61.2 and 59.3 for TNBC patients, at 1L, 2L and 3L+, respectively; there was a small meaningful difference between scores at 1L and at 2L and 3L+.

• EORTC scores were generally higher at 1L and lower at 3L+ in HR+/HER2- patients' functioning and symptoms. This differed for TNBC patients where scores were higher in 3L+ patients compared with 2L patients for physical, role, emotional, cognitive and social functioning. Meaningful differences in the scores at 2L and/or 3L+ were seen for most functioning scales, particularly in cognitive functioning and social functioning, and over half of the symptom scales in HR+/HER2- patients.

• Physicians were ‘satisfied’ or ‘very satisfied’ with current treatment for 77%, HR+/HER2- and 68% TNBC patients on 1L, 73% HR+/HER2- and 57% TNBC patients on 2L, and 59% HR+/HER2- and 54% TNBC patients on 3L+; likewise, patients' treatment satisfaction was reduced for both patient populations.

• This analysis demonstrated that patients with HR+/HER2- and triple negative ABC treated with 3L+ therapy, compared with patients treated with 1L or 2L therapies, reported a reduction in QoL, while concurrently, time to progression was shorter and overall response rate was poorer; similarly, satisfaction with treatment was reduced for both physicians and patients.

Keywords: :

• advanced breast cancer
• healthcare-related quality of life
• medical chart review
• metastatic breast cancer
• real-world
• retrospective
• treatment line
• treatment satisfaction

Breast cancer (BC) is the most common cancer occurring in women [1]. In 2020, an estimated 2.3 million women were diagnosed with BC and 684,996 women died from BC [1] worldwide. Survival depends on the stage of disease as well as by BC subtype [2]. Overall, the 5-year relative survival probability for patients with localized disease is 99.3%, reduced to 86.3% for those with regional disease and 31.0% for those with distant disease [2]. The majority of BC cases (approximately 70%) are hormone-receptor positive (HR+)/HER2- [2]. Triple-negative breast cancer (TNBC) is defined by the lack of expression of estrogen and progesterone receptors and lack of amplification of the HER2 gene [3]. TNBC accounts for about 10–15% of all BCs and is renowned for its capacity to affect younger women, metastasize early and carry poor prognosis [2,3]. Advanced breast cancer (ABC) or stage IIIb–IV disease is normally treated with systemic therapy including hormone therapy, chemotherapy, targeted therapy, immunotherapy or a combination of these [4,5]. However, none of these therapies are considered curative and most women with ABC will eventually die of their disease [6]. Treatment for HER2- ABC aims to improve survival outcomes and at least maintain quality of life (QoL) [7].

The treatment landscape for HER2- ABC is rapidly evolving, reflected in recent approvals of novel therapies and updates to clinical guidelines including the NCCN 2020 and ESMO 2020 guidelines leading to differences in availability of treatment options across regions [7,8]. There are various therapeutic approaches that could be used sequentially for HR+/HER2- and triple-negative ABC but there is limited data on clinical practice and real-world treatment outcomes. Standard endocrine therapy and a CDK4/6 inhibitor is now the preferred first-line treatment for most patients with HR+/HER2- ABC [4,9]. While CDK4/6 inhibitor therapy was added to the armamentarium for clinicians treating patients with ABC, chemotherapy is still needed for later lines [9]. Combinations of immunotherapy and chemotherapy are now emerging as the first-line standard of care for metastatic TNBC expressing PD-L1 [4]. Platinum therapy is recommended as a first like therapy for those with HER2-/BRCA mutated stage IV BC, with the PARP inhibitors a preferred option within this patient type [7]. Sequential single-agent chemotherapy is the standard of care for patients not eligible for or progressing following these targeted therapies or immunotherapy options [4,5].

Despite these treatment developments improving survival, the burden of disease in patients with ABC at later lines of therapy still remains. There is a need to understand variations in clinical practice, treatment outcomes in terms of clinical effectiveness and QoL as well as patients and physician satisfaction toward treatments for HR+/HER2- and TNBC ABC in the real-world setting, and to provide a reference point for future work. Most of the published evidence in the literature comes from the clinical trial setting. As the treatment landscape evolves, evidence from a real-world perspective adds additional context to the existing knowledge derived from the clinical trial setting, which are not always reflective of what happens in a clinical practice setting. The objective of this analysis therefore was to describe real-world treatment patterns, clinical and QoL outcomes and treatment satisfaction across treatment lines among patients with ABC in three EU countries (EU3) and USA.

Methods

Survey design

This study was an analysis of secondary data drawn from the Adelphi Advanced Breast Cancer (ABC) Disease Specific Programme (DSP^™) conducted in the EU3 (France, Italy and Spain) and USA. Data from real-world clinical practice were collected between October 2019 to April 2020.

DSPs^™ are large, multinational cross-sectional surveys with retrospective data collection of physicians and their patients presenting in a real-world clinical practice setting. The cross-sectional nature of these surveys allow for collection of data on current disease management, disease-burden impact and associated treatment effects (including QoL) at the point of most recent consultation. In addition, historic treatment data from earlier in-patient treatment journeys were retrospectively collected, using data available in patient historical clinical records. A complete description of the methods of the survey has been previously published and validated [^10-12].

Patients provided informed consent for use of their anonymized and aggregated data for research and publication in scientific journals. Data were collected in such a way that patients and physicians could not be identified directly; all data were aggregated and de-identified. Physician participation was financially incentivized, with reimbursement upon survey completion according to fair market research rates. Patients were not compensated for participation.

Data collection was undertaken in line with European Pharmaceutical Marketing Research Association [13] guidelines and as such did not require ethics committee approval. Each survey was performed in full accordance with relevant legislation at the time of data collection, including the US Health Insurance Portability and Accountability Act 1996 [14], and Health Information Technology for Economic and Clinical Health Act legislation [15].

No personal identifiable information was collected in this research. A waiver for full ethical review was granted by the Western Institutional Review Board (AG8643).

Physician & patient population

A geographically representative sample of physicians were recruited to participate in the DSP^™. Physicians were eligible to participate in this survey if they were medical oncologists, personally responsible for treatment decisions and management of patients with ABC and consulted with at least five patients with ABC a month. Patients were eligible for inclusion if they were >18 years old, had a physician-confirmed diagnosis of Stage IIIb-IV HER2- ABC, were not involved in a clinical trial at the time of data collection and had visited the physician during the study period.

Since the survey was designed to facilitate understanding of real-world clinical practice, physicians could only report data they had to hand at the time of the consultation. This represents the evidence physicians had when making any treatment and other clinical management decisions at that consultation. No tests, treatments, investigations or pre-set hypotheses were developed or performed as part of this survey.

Data collection

Physicians were instructed to complete a patient record form for the next 8–10 consecutive eligible patients with stage IIIb–IV HER2- ABC who presented for routine care. The patient record form contained detailed questions on patients' demographics and clinical characteristics and treatment satisfaction at the time of data collection, but also, retrospective data on clinical outcomes and treatment history throughout the patient's entire treatment journey. To complete the patient record form, physicians consulted existing patient clinical records and used their judgement and diagnostic skills, which is consistent with decisions made in routine clinical practice.

The clinical outcomes assessed included time to progression, overall response rate (ORR) and disease control rate (DCR), which were calculated using retrospective data from patient clinical records. Time to progression was defined as treatment duration of the patients most recently completed treatment line, where a start date was available and the physician stated that the reason for treatment discontinuation was disease progression. ORR and DCR were defined based on a modified version of the RECIST v1.1 criteria [16], with ORR including the total number of patients who had either a complete or partial response and DCR including the total number of patients who had either a complete response, partial response or stable disease. Complete response was defined as a reduction of the diameter of the largest tumor by ≥95% in size to 1 mm from current treatment initiation to time of consultation. Partial response was defined as a reduction of the diameter of the largest tumor >30% from current treatment initiation to the time of consultation. Stable disease was defined as not having a complete or partial response, as well as not having progressive disease (an increase of the diameter of the largest tumor >20% from current treatment initiation to time of consultation; minimum increase had to be >5 mm). In addition, ORR and DCR data were calculated using physician reported patient tumor sizes, and reported subjectively by physicians based on their perception of patient disease status at the end of patients' treatment, to allow for assessment of physician alignment with clinical criteria.

At consultation, physicians invited the same patients with ABC to complete a patient-reported questionnaire form. Patient-reported questionnaire forms were completed by the patient independent of their physician and were returned in a sealed envelope ensuring the patient's responses were kept confidential from their physician. The form contained questions about patients' QoL and level of satisfaction with treatment at the time of current consultation. As part of this form, patients were asked to complete three patient-reported outcome (PRO) questionnaires to further evaluate their QoL, disease burden and treatment satisfaction. PRO questionnaires were administered on completion of the patient-reported questionnaire and all questionnaires were provided in each country's native language.

PRO questionnaires included the European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire-Core 30 (EORTC QLQ-C30, version 3.0) [17] and the supplementary EORTC breast cancer-specific module (QLQ-BR23, version 1.0) [18], alongside the EuroQol-5 dimensions-5 level (EQ-5D-5L) questionnaire [19]. The Cancer Therapy Satisfaction Questionnaire (CTSQ) was used to determine treatment satisfaction [20]. All questionnaires are standardized and validated. Due to the voluntary nature of the patient-reported questionnaire, the sample size of respondents per QoL analysis varied and is reported separately for each analysis.

The 30-item EORTC QLQ-C30 consists of a global health/QoL scale, five functional scales (physical, role, cognitive, emotional and social), and nine symptom scales (fatigue, nausea/vomiting, pain, dyspnea, insomnia, appetite loss, constipation, diarrhea and financial difficulties) [17]. The 23-item EORTC QLQ-BR23 evaluates disease symptoms and treatment side effects by means of four functional scales (body image, sexual functioning, sexual enjoyment and future perspective) and four symptom scales (systemic therapy side effects, breast symptoms, arm symptoms and upset by hair loss) [18]. On both questionnaires, the global health/QoL scale ranged from 1 to 7, and each item was scored on a scale ranging from 1 to 4. A scoring algorithm was used to transform these scores to a scale ranging from 0 to 100 [21]. On the global health/QoL and functional scales, a higher score represents a better QoL and level of functioning (excluding ‘financial difficulties’ where higher scores indicate greater difficulties), respectively. On the symptom scales, a higher score represents worse symptoms.

EORTC QLQ-C30 results were reported in terms of meaningful minimum differences, which are defined differently per subscale of the EORTC QLQ-C30 (Supplementary Table 1; as detailed in reference [22]. All differences in EORTC QLQ-C30 scores that met the minimum threshold to be considered a small difference were reported [22]. Meaningful minimal differences in EORTC QLQ-BR23 scores were reported if they were ≥5, therefore meeting the minimum threshold defined for a small difference (as detailed in reference [23]).

The EQ-5D-5L is a generic instrument used to assess QoL across a variety of health conditions and treatments [19,24]. It includes both a descriptive system of the respondent's health state and a visual analog scale (EQ-VAS). The EQ-5D-5L descriptive system incorporates five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Each dimension has five levels ranging from 1 (no problems) to 5 (extreme problems). Patient responses to the EQ-5D-5L descriptive dimensions were combined to give an EQ-5D health state/profile for each patient. Weights derived from the population of each country were then applied to the health states, converting them to a preference-weighted summary score (EQ-5D health utility index), where a score of 1 indicates perfect health, and 0 indicates a state equivalent to death. The EQ-VAS represents self-rated overall health on a scale from 0 (“the worst health you can imagine”) to 100 (“the best health you can imagine”) [19,24]. For both EQ-5D-5L and EQ-VAS, higher scores indicate better QoL.

The CTSQ is a 16-item questionnaire evaluating three domains related to patients' satisfaction with cancer therapy (expectations of therapy, feelings about side effects, and satisfaction with therapy) [20]. The ‘expectations of therapy’ domain comprises five items; the ‘feelings about side effects’ domain comprises four items; and the ‘satisfaction with therapy’ domain comprises seven items. All 16 CTSQ items are expressed on a scale ranging from 0 to 100 for each CTSQ domain, with higher scores associated with better outcomes on each domain.

Analysis

Patient data were stratified by line of current therapy. Data were summarized using descriptive analyses and were reported at a regional level (i.e., USA and aggregated EU3 data). Means and standard deviations (SD) were calculated for continuous variables (number of observations), and frequency and percentages were calculated for categorical variables (patient numbers). No statistical hypothesis testing was performed. Meaningful minimal differences were reported for the EORTC QLQ-C30 and EORTC QLQ-BR23. HR+/HER2- and TNBC patient demographics were analyzed separately, and patients with unknown hormone-receptor (HR) status were not reported in the results.

In summarizing treatments found in the ABC DSP, data were subsequently categorized into the following five treatment groupings which reflect treatment guideline recommendations: chemotherapy monotherapy, hormone therapy monotherapy, hormone plus targeted therapy, any immunotherapy and other treatments.

Chemotherapy monotherapy was defined as the use of a single chemotherapy agent, including capecitabine, carboplatin, cyclophosphamide, docetaxel, doxorubicin, doxorubicin liposomal, epirubicin, eribulin, etoposide, 5-fluorouracil, gemcitabine, ixqabepilone, methotrexate, mitoxantrone, nab paclitaxel, paclitaxel, vinorelbine and other chemotherapy. Hormone therapy monotherapy was defined as the use of a single hormone therapy, including anastrozole, exemestane, fulvestrant, goserelin, letrozole, leuprorelin, raloxifene, tamoxifen and toremifene. Hormone plus targeted therapy was the combined use of a hormone therapy (as listed above) and targeted therapy (abemaciclib, alpelisib bevacizumab, everolimus, olaparib, palbociclib and ribociclib). Any immunotherapy was use of an immunotherapy (atezolizumab, pembrolizumab) irrespective of other treatments used, and ‘other’ therapies included any treatment combinations not defined above.

Results

Physician & patient population

A flowchart is shown in Figure 1, indicating the number of patients identified and included within this analysis. A total of 232 oncologists were recruited for the ABC DSP^™ (57 US and 175 EU3 [64 France, 55 Italy, 56 Spain]). A total of 1977 patients in two patient populations, consisting of 1641 HR+/HER2- patients (1316 EU3 [403 Spain; 462 France; 451 Italy] and 325 US) and 264 patients with TNBC (182 EU3 [54 Spain; 63 France; 65 Italy] and 82 USA) were included. A total of 72 patients had unknown HR status, and were excluded from this analysis. Patient-reported questionnaire forms were completed by a total of 597 patients (101 USA and 496 EU3 [142 France, 106 Italy, 248 Spain]). The sample size varied with treatment line and outcome. Due to the low base size of the TNBC patient population, this data was not analyzed in detail, but was presented as supplementary tables for reference (Supplementary Tables 2 & 3).

Figure 1. Flowchart of the number of patients included and identified within the analysis.

ABC: Advanced breast cancer; HR: Hormone-receptor; PRF: Patient Record Form; TNBC: Triple-negative breast cancer.

¹Data provided in supplementary tables.

²Data not provided.

At the time of data collection, 909 (46%) of the overall patient population were currently on first-line treatment (1L) (757 [47%] HR+/HER2-; 120 [46%] TNBC), 767 (39%) were currently on second-line treatment (2L) (642 [39%] HR+/HER2-; 97 [37%] TNBC), 285 (15%) were currently on third- or later-lines of treatment (3L+) (227 [14%] with HR+/HER2-, and 46 [17%] with TNBC) and 16 (1%) of patients had received an unknown number of treatment lines (15 [1%] with HR+/HER2-, and 1 with TNBC).

Demographic & clinical characteristics

Demographic and clinical characteristics of patients are shown in Tables 1 & 2.

Table 1. Baseline demographics among adult patients with HR+/HER2- ABC.

	Total (n = 1641)			EU3 (n = 1316)			USA (n = 325)
	1L (n = 757)	2L (n = 642)	3L+ (n = 227)	1L (n = 586)	2L (n = 519)	3L+ (n = 202)	1L (n = 171)	2L (n = 123)	3L+ (n = 25)
Age (years)	64.1	65.3	64.4	63.7	65.4	64.2	65.5	65.2	65.9
Female, n (%)	750 (99)	638 (99)	225 (99)	583 (99)	515 (99)	200 (99)	167 (98)	123 (100)	25 (100)
Ethnicity, n (%)	n = 757	n = 642	n = 227	n = 586	n = 519	n = 202	n = 171	n = 123	n = 25
White/Caucasian	679 (90)	564 (88)	204 (90)	560 (96)	484 (93)	189 (94)	119 (70)	80 (65)	15 (60)
African–American (USA only)	35 (5)	30 (5)	7 (3)	0 (0)	0 (0)	0 (0)	35 (20)	30 (24)	7 (28)
Asian-Indian subcontinent	4 (1)	1 (0)	1 (0)	0 (0)	1 (0)	0 (0)	4 (2)	0 (0)	1 (4)
Hispanic/Latino	16 (2)	23 (4)	5 (2)	10 (2)	16 (3)	5 (2)	6 (4)	7 (6)	0 (0)
Afro-Caribbean (EU only)	7 (1)	6 (1)	1 (0)	7 (1)	6 (1)	1 (0)	0 (0)	0 (0)	0 (0)
Family history of BRCA-driven cancers, n (%)	n = 716	n = 603	n = 204	n = 564	n = 501	n = 186	n = 152	n = 102	n = 18
Patient has a family history	93 (13)	95 (16)	25 (12)	71 (13)	79 (16)	22 (12)	22 (14)	16 (16)	3 (17)
Patient does not have a family history	623 (87)	508 (84)	179 (88)	493 (87)	422 (84)	164 (88)	130 (86)	86 (84)	15 (83)

15 patients with HR+/HER2- ABC had received an unknown number of treatment lines.

TNBC population size was limited and therefore presented as supplementary tables.

1L: First-line therapy; 2L: Second-line therapy; 3L+: Third- and later-line of therapy; ABC: Advanced breast cancer; EU3: European Union three countries (France, Italy and Spain); TNBC: Triple-negative breast cancer.

Table 2. Baseline clinical characteristics among adult patients with HR+/HER2- ABC.

	Total			EU3			USA
	1L (n = 757)	2L (n = 642)	3L+ (n = 227)	1L (n = 586)	2L (n = 519)	3L+ (n = 202)	1L (n = 171)	2L (n = 123)	3L+ (n = 25)
Menopausal status, females, n (%)	n = 750	n = 638	n = 225	n = 583	n = 515	n = 200	n = 167	n = 123	n = 25
Post-menopausal	656 (87)	589 (92)	210 (93)	505 (87)	479 (93)	185 (93)	151 (90)	110 (89)	25 (100)
Pre-menopausal	48 (6)	23 (4)	8 (4)	43 (7)	15 (3)	8 (4)	5 (3)	8 (7)	0 (0)
Peri-menopausal	46 (6)	26 (4)	7 (3)	35 (6)	21 (4)	7 (4)	11 (7)	5 (4)	0 (0)
Stage at time of data collection, n (%)	n = 757	n = 642	n = 227	n = 586	n = 519	n = 202	n = 171	n = 123	n = 25
Stage IIIb	43 (6)	9 (1)	2 (1)	29 (5)	6 (1)	2 (1)	14 (8)	3 (2)	0 (0)
Stage IIIc	18 (2)	9 (1)	2 (1)	14 (2)	7 (1)	2 (1)	4 (2)	2 (2)	0 (0)
Stage IV	696 (92)	624 (97)	223 (98)	543 (93)	506 (97)	198 (98)	153 (89)	118 (96)	25 (100)
ECOG scores at time of data collection, n (%)^†	n = 757	n = 642	n = 227	n = 586	n = 519	n = 202	n = 171	n = 123	n = 25
0	258 (34)	168 (26)	36 (16)	225 (38)	143 (28)	32 (16)	33 (19)	25 (20)	4 (16)
1	405 (54)	342 (53)	115 (51)	293 (50)	279 (54)	102 (50)	112 (65)	63 (51)	13 (52)
2	68 (9)	89 (14)	50 (22)	50 (9)	67 (13)	46 (23)	18 (11)	22 (18)	4 (16)
3	14 (2)	18 (3)	16 (7)	10 (2)	14 (3)	14 (7)	4 (2)	4 (3)	2 (8)
4	12 (2)	23 (4)	10 (4)	8 (1)	14 (3)	8 (4)	4 (2)	9 (7)	2 (8)
Visceral status, metastatic HER2- patients, n (%)	n = 696	n = 624	n = 223	n = 543	n = 506	n = 198	n = 153	n = 118	n = 25
Visceral metastasis	291 (42)	307 (49)	135 (61)	240 (44)	258 (51)	125 (63)	51 (33)	49 (42)	10 (40)
Non-visceral metastasis	395 (57)	309 (50)	87 (39)	293 (54)	241 (48)	72 (36)	102 (67)	68 (58)	15 (60)
Number of metastatic sites, n (%)	n = 696	n = 624	n = 223	n = 543	n = 506	n = 198	n = 153	n = 118	n = 25
1 site	348 (50)	241 (39)	71 (32)	252 (46)	186 (37)	58 (29)	96 (63)	55 (47)	13 (52)
2 sites	230 (33)	233 (37)	82 (37)	189 (35)	197 (39)	74 (37)	41 (27)	36 (31)	8 (32)
3+ sites	118 (17)	150 (24)	70 (31)	102 (19)	123 (24)	66 (33)	16 (10)	27 (23)	4 (16)
Bone-only metastases, metastatic HER2- patients, n (%)	n = 696	n = 624	n = 223	n = 543	n = 506	n = 198	n = 153	n = 118	n = 25
Patient has bone only metastases	245 (35)	196 (31)	50 (22)	176 (32)	149 (29)	37 (19)	69 (45)	47 (40)	13 (52)
Time since ABC diagnosis (days)	n = 713	n = 584	n = 208	n = 561	n = 482	n = 188	n = 152	n = 102	n = 20
Days since ABC diagnosis	270.9	720.6	1151.0	273.8	738.5	1170.7	260.1	636.4	965.9
Type of current therapy, n (%)	n = 757	n = 642	n = 227	n = 586	n = 519	n = 202	n = 171	n = 123	n = 25
Chemotherapy monotherapy	44 (6)	122 (19)	89 (39)	37 (6)	110 (21)	86 (43)	7 (4)	12 (10)	3 (12)
Hormone monotherapy	131 (17)	131 (20)	29 (13)	95 (16)	99 (19)	23 (11)	36 (21)	32 (26)	6 (24)
Hormone plus targeted therapy	468 (62)	289 (45)	69 (30)	376 (64)	240 (46)	58 (29)	92 (54)	49 (40)	11 (44)
Any immunotherapy-based therapy	5 (1)	8 (1)	2 (1)	2 (0)	2 (0)	0 (0)	3 (2)	6 (5)	2 (8)
Other	109 (14)	92 (14)	38 (17)	76 (13)	68 (13)	35 (17)	33 (19)	24 (20)	3 (12)

† ECOG performance status: score of 1, fully active, no performance restrictions; 2, strenuous physical activity restricted, fully ambulatory and able to carry out light work; 3, capable of all self-care but unable to carry out any work activities up and about >50% of waking hours; 4, capable of only limited self-care, confined to bed or chair >50% of waking hours; 5, completely disabled, cannot carry out any self-care, totally confined to bed or chair.

TNBC population size was limited and therefore presented as supplementary tables.

1L: First-line therapy; 2L: Second-line therapy; 3L+: Third- and later-line of therapy; ABC: Advanced breast cancer; ECOG: Eastern Cooperative Oncology Group; EU3: European Union three countries (France, Italy and Spain); TNBC: Triple-negative breast cancer.

Overall, across all treatment lines, patients had a mean age of 63.5 years, HR+/HER2- patients were older than TNBC patients (64.6 and 57.5, respectively), at the time of data collection, 99% were females and most were White/Caucasian. Generally, the proportion of patients who were currently at 3L+ with an Eastern Cooperative Oncology Group (ECOG) score of 0 were lower in comparison to patients currently at 1L, while the proportion of patients with an ECOG score of 2 or higher was observed to be highest in patients currently at 3L+.

At the time of data collection, most patients had progressed to stage IV (95% with HR+/HER2- and 88% with TNBC), with the highest proportions observed in patients currently at 2L or 3L+. Overall >80% of the population were HR+/HER2- across all treatment lines, while the remaining patients had TNBC.

Less than half of patients currently at 1L with HR+/HER2- (42%) had visceral metastases, however this proportion was higher in patients currently at 3L+ at 61%. Conversely, 82% of TNBC patients at 1L had visceral metastases, with lower proportions of visceral metastases observed in patients at 2L and 3L+ (80 and 70%, respectively).

Mean time since ABC diagnosis for patients currently at 1L, 2L and 3L+ was 271, 721 and 1151 days, respectively, for HR+/HER2- patients, and 202, 476 and 668 days, respectively, for TNBC patients. For EU3 patients, mean time since ABC diagnosis for patients at 1L, 2L and 3L+ was 274, 739 and 1171 days, respectively, for HR+/HER2- patients, and 225, 516 and 709 days, respectively for TNBC patients. For USA patients, mean time since ABC diagnosis for patients at 1L, 2L and 3L+ was 260, 636 and 966 days, respectively, for HR+/HER2- patients, and 159, 331 and 488 days, respectively, for TNBC patients.

Current treatment & response

Current treatment is shown in Table 2. Overall, hormone plus targeted therapy combinations were the most frequently prescribed in patients with HR+/HER2- currently at 1L and 2L (62 and 45%, respectively). Chemotherapy was most frequently prescribed in patients with HR+/HER2- currently at 3L+ (39%), while few patients with HR+/HER2- currently at 1L received chemotherapy (6%). Hormone monotherapy was prescribed mostly in patients with HR+/HER2- currently at 2L (20%). Immunotherapy-based regimens were prescribed for very few HR+/HER2- patients irrespective of current line of therapy (1%).

Chemotherapy was the most frequently prescribed in patients with TNBC currently at 2L and 3L+ (51 and 48%, respectively), while other treatments were more commonly prescribed in patients currently at 1L (53%). Hormone monotherapy and hormone plus targeted therapy combinations were prescribed for very few TNBC patients across all lines of therapy (≥3 and ≥9%, respectively). Immunotherapy-based regimens were prescribed mostly in 1L (17%) patients with TNBC, with chemotherapy monotherapy prescribed for 24% of 1L patients with TNBC.

Current response to treatment is shown in Table 3. The median time to progression on 2L, 3L and 4L was 13.8, 11.0 and 5.9 months, respectively for patients with HR+/HER2-, whereas mean time to progression on 2L, 3L and 4L was less for patients with TNBC, 8.0, 4.6 and 3.8 months, respectively. Time to progression was shorter in 3L+ patients compared with 2L patients for both EU3 and USA patients in both patient populations. The ORR of all 1L, 2L and 3L+ patients was 54, 53 and 42%, respectively, as calculated based on modified RECIST v1.1 criteria [25]. Generally, ORR was highest in 1L and 2L EU3 HR+/HER2- patients and in 1L and 3L+ EU3 TNBC patients. ORR was highest in USA patients on later lines for both patient populations. The DCR of all 1L, 2L and 3L+ patients was 99, 89 and 86%, respectively. The DCR was 99, 90 and 84% In 1L, 2L and 3L+ HR+/HER2- patients, respectively, and 98, 82 and 94% in 1L, 2L and 3L+ TNBC patients, respectively. For HR+/HER2- patients, physician-reported ORR and DCR were lower than, but followed a similar pattern to, those calculated based on modified RECIST v1.1 criteria. For TNBC patients, physician-reported DCR were lower than those calculated based on modified RECIST v1.1 criteria, however physician-reported ORR was higher in 1L and 3L+ patients.

Table 3. Time to progression and treatment outcomes among adult patients with HR+/HER2- ABC.

	Total				EU3				USA
	1L (n = 757)	2L (n = 642)	3L (n = 208)	4L (n = 19)	1L (n = 586)	2L (n = 519)	3L (n = 186)	4L (n = 16)	1L (n = 171)	2L (n = 123)	3L (n = 22)	4L^‡ (n = 3)
Time to progression on previous therapy	–	n = 452	n = 141	n = 10	–	n = 380	n = 130	n = 10	–	n = 72	n = 11	n = 0
Time to progression (months) [IQR]	–	13.8 [8.1–13.8]	11.0 [6.3–16.4]	5.9 [3.5–7.3]	–	14.0 [9.0–14.0]	11.0 [6.7–16.5]	5.9 [3.5–7.3]	–	10.7 [5.2–10.7]	10.9 [5.9–15.9]	0 [0]
ORR and DCR of current therapy based on RECIST v1.1, n (%)^†	n = 237	n = 228	n = 69	n/a	n = 183	n = 188	n = 63	n/a	n = 54	n = 40	n = 6	n/a
ORR	130 (55)	125 (55)	28 (41)	n/a	107 (58)	97 (52)	25 (40)	n/a	23 (43)	28 (70)	3 (50)	n/a
DCR	235 (99)	205 (90)	58 (84)	n/a	181 (99)	169 (90)	52 (83)	n/a	54 (100)	36 (90)	6 (100)	n/a
Physician-reported ORR and DCR of current therapy, n (%)^†	n = 757	n = 642	n = 227	n/a	n = 586	n = 519	n = 202	n/a	n = 171	n = 123	n = 25	n/a
ORR	418 (55)	271 (42)	73 (32)	n/a	338 (58)	210 (40)	66 (33)	n/a	80 (47)	61 (50)	7 (28)	n/a
DCR	572 (76)	470 (73)	169 (74)	n/a	473 (81)	389 (75)	156 (77)	n/a	99 (58)	81 (66)	13 (52)	n/a

† RECIST v1.1. [38].

‡ Low sample size.

TNBC population size was limited and therefore presented as supplementary tables.

1L: First-line therapy; 2L: Second-line therapy; 3L: Third-line therapy; 4L: Fourth-line therapy; ABC: Advanced breast cancer; DCR: Disease control rate; EU3: European Union three countries (France, Italy and Spain); n/a: Not applicable; ORR: Objective response rate; TNBC: Triple-negative breast cancer.

Current treatment outcomes in QoL

QoL is detailed in Tables 4 & 5. The average EQ-5D-5L health utility score of HR+/HER2- patients was observed to be lower in patients currently at later lines of treatment, ranging from 0.77 in patients currently at 1L, through 0.70 in patients at 2L, to 0.65 in patients at 3L+. Similarly, EQ-5D-5L VAS mean scores were 70.9, 67.4 and 63.4 in patients currently receiving 1L, 2L and 3L+, respectively. A similar trend across patients receiving later therapy lines was also observed in HR+/HER2- EU3 patients, however, scores remained more consistent across therapy lines in HR+/HER2- USA patients. This trend differed for TNBC patients, as both the global and EU3 EQ-5D-5L health utility and EQ-5D-5L VAS scores were higher in patients currently at 3L+ compared with those on 2L. EQ-5D-5L health utility scores for TNBC USA patients were highest in those currently at 3L+ (1.00) and lowest in those at 1L (0.75).

Table 4. EQ-5D-5L health utility and VAS scores and EORTC QLQ-C30 functional scale scores among adult patients with HR+/HER2- ABC.

	Total			EU3			USA
	1L	2L	3L+	1L	2L	3L+	1L	2L	3L+
EQ-5D-5L health utility^†	n = 244	n = 205	n = 50	n = 201	n = 164	n = 48	n = 43	n = 41	n = 2
Mean score	0.77	0.70	0.65	0.76	0.68	0.65	0.82	0.77	0.71
EQ-5D-5L VAS^‡	n = 250	n = 207	n = 50	n = 205	n = 166	n = 48	n = 45	n = 41	n = 2
Mean score	70.9	67.4	63.4	70.1	66.1	62.8	74.3	72.7	77.5
EORTC QLQ-C30: Global health status^§	n = 250	n = 205	n = 49	n = 205	n = 164	n = 47	n = 45	n = 41	n = 2
Mean score	64.6^¶^,^††	60.0^¶	57.7^††	64.1^¶^,^††	59.6^¶	57.3^††	66.9^¶	61.4^¶^,^#	66.7^#
Functional scales^§
EORTC QLQ-C30: Physical functioning	n = 250	n = 207	n = 50	n = 205	n = 166	n = 48	n = 45	n = 41	n = 2
Mean score	79.1^¶^,^††	71.3^¶	67.7^††	78.7^¶^,^††	70.9^¶^,^#	67.8^#^,^††	81.1^¶^,^††	72.8^¶^,^#	66.7^#^,^††
EORTC QLQ-C30: Role functioning	n = 250	n = 205	n = 50	n = 205	n = 164	n = 48	n = 45	n = 41	n = 2
Mean score	71.9^††	66.8^#	58.7^#^,^††	71.2^††	65.9^#	58.7^#^,^††	75.2^††	70.7^#	58.3^#^,^††
EORTC QLQ-C30: Emotional functioning	n = 250	n = 204	n = 50	n = 205	n = 163	n = 48	n = 45	n = 41	n = 2
Mean score	69.4	67.4	62.8	69.0	66.1	62.7	71.3	72.8	66.7
EORTC QLQ-C30: Cognitive functioning	n = 250	n = 204	n = 50	n = 205	n = 163	n = 48	n = 45	n = 41	n = 2
Mean score	80.2^¶^,^††	75.9^¶^,^#	68.7^#^,^††	80.2^¶^,^††	75.7^¶^,^#	68.8^#^,^††	80.0^¶^,^††	76.8^¶^,^#	66.7^#^,^††
EORTC QLQ-C30: Social functioning	n = 250	n = 204	n = 50	n = 205	n = 163	n = 48	n = 45	n = 41	n = 2
Mean score	73.8^¶^,^††	68.4^¶^,^#	57.7^#^,^††	74.1^¶^,^††	67.5^¶^,^#	59.4^#^,^††	72.6^††	72.0^#	16.7^#^,^††
EORTC QLQ-C30: Financial difficulties	n = 249	n = 204	n = 50	n = 204	n = 163	n = 48	n = 45	n = 41	n = 2
Mean score	15.1^††	18.0^#	24.0^#^,^††	13.2^††	15.7^#	22.9^#^,^††	23.7^††	26.8^#	50.0^#^,^††
EORTC QLQ-BR23: Future perspective	n = 245	n = 199	n = 49	n = 200	n = 159	n = 47	n = 45	n = 40	n = 2
Mean score	51.4^††	47.2^#	40.8^#^,^††	49.2^¶^,^††	44.2^¶	42.6^††	61.5	59.2	0.0
EORTC QLQ-BR23: Body image	n = 248	n = 201	n = 49	n = 203	n = 161	n = 47	n = 45	n = 40	n = 2
Mean score	73.7^††	68.8^#	63.4^#^,^††	71.3^††	67.1	66.1^††	84.3^¶	75.4^¶	0.0
EORTC QLQ-BR23: Sexual functioning	n = 239	n = 190	n = 47	n = 197	n = 151	n = 45	n = 42	n = 39	n = 2
Mean score	14.9	15.3^#	10.6^#	14.0	13.7	11.1	19.0	21.4	0.0
EORTC QLQ-BR23: Sexual enjoyment	n = 132	n = 105	n = 20	n = 107	n = 83	n = 20	n = 25	n = 22	n = -
Mean score	28.3^††	27.0^#	21.7^#^,^††	28.7^¶^,^††	23.7^¶	21.7^††	26.7^¶	39.4^¶	0.0

† For EQ-5D-5L, higher scores indicate better health status.

‡ For EQ-VAS, higher scores indicate better health status.

§ For the EORTC QLQ-C30 Global Health Status score and the EORTC-QLQ-C30 and -BR23 functional scale scores, higher scores indicate better functioning, excluding ‘Financial difficulties’ where higher scores indicate greater difficulties.

¶ Meaningful difference between the 1L and 2L results for this EORTC score.

# Meaningful difference between the 2L and 3L results for this EORTC score.

†† Meaningful difference between the 1L and 3L results for this EORTC score.

The number of USA patients reporting results at 3L+ is too small to meaningfully evaluate.

A difference in EORTC QLQ-C30 score was considered meaningful when scores were: ≥3 for cognitive functioning, financial difficulties, nausea and vomiting and diarrhea; ≥4 for global health status, dyspnea and insomnia; ≥5 for physical functioning, social functioning, fatigue, appetite loss and constipation; and ≥6 for role functioning and pain; a meaningful difference for the EORTC QLQ C30 score of emotional functioning could not be estimated [22A difference in EORTC QLQ-BR23 score was considered meaningful when scores were: <5, no change; 5–10, small change; >10–20, moderate change; and >20, large change [23].

TNBC population size was limited and therefore presented as supplementary tables.

1L: First-line therapy; 2L: Second-line therapy; 3L+: Third- and later-line of therapy; ABC: Advanced breast cancer; EORTC: European Organization for Research and Treatment of Cancer; EQ-5D-5L: EuroQol-5 dimensions-5 level; EU3: European Union three countries (France, Italy and Spain); QLQ-BR23: Quality of Life Questionnaire-Breast cancer-specific module 23 item; QLQ-C30: Quality of Life Questionnaire-Core 30; TNBC: Triple-negative breast cancer; VAS: Visual analog scale.

Table 5. EORTC QLQ-C30 symptom scale scores among adult patients with HR+/HER2- ABC.

	Total			EU3			USA
	1L	2L	3L+	1L	2L	3L+	1L	2L	3L+
Symptom scales^†
EORTC QLQ-C30: Fatigue	n = 250	n = 206	n = 50	n = 205	n = 165	n = 48	n = 45	n = 41	n = 2
Mean score	32.0^‡^,^¶	38.1^‡^,^§	43.7^§^,^¶	32.1^‡^,^¶	39.7^‡	44.6^¶	31.6^¶	31.7^§	22.2^§^,^¶
EORTC QLQ-C30: Nausea and vomiting	n = 249	n = 206	n = 50	n = 204	n = 165	n = 48	n = 45	n = 41	n = 2
Mean score	12.7^¶	14.2	16.7^¶	12.8^‡^,^¶	16.1^‡	17.4^¶	12.2^‡	6.9^‡	0.0
EORTC QLQ-C30: Pain	n = 250	n = 206	n = 50	n = 205	n = 165	n = 48	n = 45	n = 41	n = 2
Mean score	26.9^‡^,^¶	33.6^‡	34.0^¶	27.5^‡^,^¶	35.3^‡	34.4^¶	24.1	26.8	25.0
EORTC QLQ-C30: Dyspnoea	n = 249	n = 203	n = 50	n = 204	n = 163	n = 48	n = 45	n = 40	n = 2
Mean score	15.8^‡^,^¶	20.0^‡^,^§	26.7^§^,^¶	17.2^‡^,^¶	21.5^‡^,^§	27.1^§^,^¶	9.6^‡^,^¶	14.2^‡	16.7^¶
EORTC QLQ-C30: Insomnia	n = 249	n = 206	n = 50	n = 204	n = 165	n = 48	n = 45	n = 41	n = 2
Mean score	28.4^¶	31.7^§	40.0^§^,^¶	29.4^¶	31.9^§	41.7^§^,^¶	23.7^‡	30.9^‡	0.0
EORTC QLQ-C30: Appetite loss	n = 249	n = 205	n = 50	n = 204	n = 164	n = 48	n = 45	n = 41	n = 2
Mean score	22.0^¶	25.5^§	34.7^§^,^¶	21.6^¶	26.0^§	34.7^§^,^¶	23.7^¶	23.6^§	33.3^§^,^¶
EORTC QLQ-C30: Constipation	n = 248	n = 206	n = 49	n = 203	n = 165	n = 47	n = 45	n = 41	n = 2
Mean score	11.7	14.9	16.3	13.5	16.0	17.0	3.7^‡	10.6^‡	0.0
EORTC QLQ-C30: Diarrhea	n = 250	n = 203	n = 50	n = 205	n = 162	n = 48	n = 45	n = 41	n = 2
Mean score	12.8	13.5	15.3	13.7	15.6	16.0	8.9^‡	4.9^‡	0.0
EORTC QLQ-BR23: Systemic therapy side effects	n = 250	n = 205	n = 50	n = 205	n = 165	n = 48	n = 45	n = 40	n = 2
Mean score	19.2^¶	22.8^§	28.0^§^,^¶	20.3^¶	24.6	28.5^¶	14.2	15.5	16.7
EORTC QLQ-BR23: Breast symptoms	n = 247	n = 205	n = 49	n = 202	n = 164	n = 47	n = 45	n = 41	n = 2
Mean score	15.8	15.0	16.7	17.1	15.4	17.4	10.1	13.2	0.0
EORTC QLQ-BR23: Arm symptoms	n = 247	n = 205	n = 48	n = 202	n = 164	n = 46	n = 45	n = 41	n = 2
Mean score	15.0	17.8	19.9	16.0	18.5	20.8	10.5	14.9	0.0
EORTC QLQ-BR23: Upset by hair loss	n = 145	n = 124	n = 41	n = 116	n = 104	n = 41	n = 29	n = 20	n = -
Mean score	32.2	34.9	32.5	34.2	36.5	32.5	24.1	26.7	0.0

† For EORTC-QLQ-C30 and -BR23 symptom scale scores, higher scores indicate worse symptoms.

‡ Meaningful difference between the 1L and 2L results for this EORTC score.

§ Meaningful difference between the 2L and 3L results for this EORTC score.

¶ Meaningful difference between the 1L and 3L results for this EORTC score.

The number of USA patients reporting results at 3L+ is too small to meaningfully evaluate.

A difference in EORTC QLQ-C30 score was considered meaningful when scores were: ≥3 for cognitive functioning, financial difficulties, nausea and vomiting, and diarrhea; ≥4 for global health status, dyspnea, and insomnia; ≥5 for physical functioning, social functioning, fatigue, appetite loss, and constipation; and ≥6 for role functioning, and pain; a meaningful difference for the EORTC QLQ C30 score of emotional functioning could not be estimated [22A difference in EORTC QLQ-BR23 score was considered meaningful when scores were: <5, no change; 5–10, small change; >10–20, moderate change; and >20, large change [23].

TNBC population size was limited and therefore presented as supplementary tables.

1L: first-line therapy; 2L: second-line therapy; 3L+: third and later lines of therapy; ABC: advanced breast cancer; EORTC: European Organization for Research and Treatment of Cancer; EU3: European Union three countries (France, Italy, and Spain); QLQ-BR23: Quality of Life Questionnaire-Breast cancer-specific module 23 items; QLQ-C30: Quality of Life Questionnaire-Core 30; TNBC: Triple-negative breast cancer.

There were minimal small meaningful differences in the EORTC QLQ-C30 global health status mean score for both patient populations (i.e. differences in scores of 3–6) across patients currently at 1L, 2L and 3L+ (HR+/HER2-; 64.6, 60.0 and 57.7 and TNBC; 64.0, 61.2 and 59.3, respectively). In HR+/HER2- patients, EORTC scores were generally higher in patients currently at 1L and lower in patients at 3L+ for functioning and symptoms. This differed for TNBC patients, where scores were higher in patients currently at 3L+ for physical, role, emotional, cognitive, and social functioning. Compared with patients currently at 1L, meaningful minimum differences in the scores for patients at 2L and/or 3L+ were seen for most functioning scales in HR+/HER2- patients, particularly in cognitive functioning and social functioning, and over half of the symptom scales, particularly in insomnia and dyspnoea. Meaningful minimum differences in the scores for patients currently at 2L and/or 3L+ were seen in most symptom scores in TNBC patients, particularly in insomnia and pain. Differences between patients at 1L, 2L and 3L+ were trivial for nausea/vomiting, breast symptoms, and arm symptoms in both patient populations.

EORTC scores of both EU3 and USA patients were generally higher in patients currently at 1L and lower in patients currently at 3L+ in functioning and symptoms, regardless of patient type. Considering EU3 HR+/HER2- and TNBC patients only, compared with patients currently receiving 1L, those receiving 2L had a worse global health status, experienced poorer physical, role, cognitive, and emotional functioning, experienced worse fatigue, pain, dyspnoea, insomnia, diarrhoea and had greater financial difficulties. EU3 HR+/HER2- patients currently at 3L+ specifically experienced worse insomnia and had greater appetite loss and financial difficulties compared with those at 2L. EU3 HR+/HER2- patients showed no meaningful differences between treatment lines in sexual functioning, constipation, diarrhoea, breast symptoms and ‘upset by hair loss’, while EU3 TNBC patients showed no meaningful differences between treatment lines in nausea and vomiting only.

HR+/HER2- USA patients currently at 2L had a worse global health status, and experienced poorer physical, role, and cognitive functioning as well as worse, financial difficulties, dyspnoea, insomnia, and constipation, than USA patients at 1L. TNBC patients scores at 3L+ were better for physical, role, emotional, cognitive, and social functioning, and additionally reported fewer financial difficulties, and better fatigue, nausea and vomiting, pain, dyspnoea, insomnia, appetite loss and constipation compared with patients currently at 2L. HR+/HER2- USA patients at 2L also had a worse body image and future perspective than USA patients at 1L (Table 4).

There were no meaningful differences between 1L and 2L USA HR+/HER2- patients with regards to emotional, social, and sexual functioning, future perspective, or in fatigue, pain, appetite loss, systemic therapy side effects, and ‘upset by hair loss’. The base size for US TNBC patients was too low to allow for comparison between lines of therapy, and similarly due to low sample size of USA 3L+ patients in all patient populations, limited interpretation, and analysis of patient QoL could be made for these patient groups.

Physician & patient satisfaction with current treatment

Treatment satisfaction is shown in Table 6.

Table 6. Physician and patient satisfaction with current treatment among adult patients with HR+/HER2- ABC.

	Total			EU3			USA
	1L	2L	3L+	1L	2L	3L+	1L	2L	3L+
Physician's satisfaction with patient's current treatment, n (%)	n = 757	n = 642	n = 227	n = 586	n = 519	n = 202	n = 171	n = 123	n = 25
Very dissatisfied	2 (0)	1 (0)	1 (0)	1 (0)	1 (0)	1 (0)	1 (1)	0 (0)	0 (0)
Dissatisfied	12 (2)	7 (1)	3 (1)	8 (1)	4 (1)	3 (1)	4 (2)	3 (2)	0 (0)
Moderately satisfied	158 (21)	166 (26)	88 (39)	125 (21)	156 (30)	82 (41)	33 (19)	10 (8)	6 (24)
Satisfied	400 (53)	364 (57)	118 (52)	320 (55)	284 (55)	101 (50)	80 (47)	80 (65)	17 (68)
Very satisfied	185 (24)	104 (16)	17 (7)	132 (23)	74 (14)	15 (7)	53 (31)	30 (24)	2 (8)
CTSQ^†,‡
Satisfaction with therapy	n = 247	n = 204	n = 50	n = 203	n = 163	n = 48	n = 44	n = 41	n = 2
Mean Score	78.0	72.4	67.2	77.3	70.3	66.2	80.8	80.6	92.9
Feelings about side effects	n = 243	n = 200	n = 50	n = 199	n = 159	n = 48	n = 44	n = 41	n = 2
Mean score	63.3	58.4	54.3	63.1	56.4	53.5	64.2	66.0	71.9
Expectations of therapy	n = 248	n = 202	n = 50	n = 203	n = 161	n = 48	n = 45	n = 41	n = 2
Mean Score^‡	69.3	66.4	56.2	67.4	64.4	55.6	77.9	74.6	70.0

† For the CTSQ scale scores, higher scores indicate better outcomes.

‡ Reported on patient self-completion questionnaires.

TNBC population size was limited and therefore presented as supplementary tables.

1L: First-line therapy; 2L: Second-line therapy; 3L+: Third- and later-line of therapy; ABC: Advanced breast cancer; CTSQ: Cancer Therapy Satisfaction Questionnaire; EU3: European Union three countries (France, Italy and Spain); TNBC: Triple-negative breast cancer.

Physicians were ‘satisfied’ or ‘very satisfied’ with current treatment for 76% of all patients at 1L, 73% of all patients at 2L, and 59% of all patients at 3L+, with similar proportions were observed for the HR+/HER2- patient population (77% 1L; 73% 2L; 59% 3L+). Generally, a lower proportion of physicians were ‘satisfied’ or ‘very satisfied’ with current treatment for TNBC patients at each line (68% 1L; 57% 2L; 54% 3L+). Similar proportions of EU3 patients reported being ‘satisfied’ with current treatment regardless of patient type or line of current therapy (HR+/HER2-, 55% 1L, 55% 2L, 50% 3L+; TNBC, 47% 1L, 48% 2L, 42% 3L+). Forty-seven percent of all USA physicians completing questionnaires for HR+/HER2- patients reported being satisfied with current 1L treatment, a higher percentage reported being satisfied with current 2L and 3L treatments (65 and 68%, respectively). This was different for TNBC patients 67% of US physicians reported being satisfied with current 1L treatment, lower percentages were observed in patients currently at 2L and 3L+ (63 and 54%, respectively).

Overall, for all HR+/HER2- patients, the patient-reported mean CTSQ domain scores for satisfaction with therapy, feelings about side effects, and expectations of therapy were reported to be lower for patients at 1L compared with patients at 2L and 3L+. A similar trend for satisfaction with therapy and expectations of therapy scores were observed with TNBC patients, however, with this patient population conversely feelings about side effects were better in patients on 3L+ compared with 1L patients. Mean satisfaction with therapy scores for HR+/HER2- patients were 78.0, 72.4, and 67.2 for 1L, 2L, and 3L+ patients, respectively, and 72.3 on 1L, 70.2 on 2L and 63.8 on 3L+, respectively for TNBC patients.

Discussion

This real-world study of patients with Stages IIIb-IV, HR+/HER2- and TNBC ABC showed that despite recent treatment developments, the burden of disease at later lines of therapy still remains. In comparison to patients currently receiving initial treatment for ABC, QoL and treatment satisfaction was lower in patients currently at later lines of treatment (2L TNBC patients and 3L+ HR+/HER2- patients). Concurrently, overall, median time to progression was shorter and ORR and DCR were poorer in both 1L patient populations compared with those patients on 2L and 3L. While other real-world and controlled studies have documented QoL and treatment satisfaction in patients with ABC before and after treatment with selected therapies [^26-31], to our knowledge, studies which have reported QoL and treatment satisfaction between therapy lines in real-world setting are limited, with this study adding to the existing evidence base.

Treatment was generally as per clinical guidelines [7,8], with standard endocrine therapy and a targeted therapy the preferred 1L therapy [4,9]. Thus, at the time of data collection HR+/HER2- patients mostly received hormone therapy plus targeted therapy combinations at 1L and 2L, particularly in the EU3, and chemotherapy at 3L+. TNBC patients mostly received chemotherapy at 2L and 3L+ while other treatments were the most frequently prescribed at 1L. Whilst immunotherapy-based regimens were prescribed for patients with TNBC as a first-line therapy, the proportion may be lower than expected due to data collection occurring just after the approval date for these treatments [4].

Our study demonstrated worse functioning and symptoms in HR+/HER2- patients who were receiving later lines of treatment, compared with patients on their initial 1L treatment. For TNBC patients, functioning scores were higher in patients at 3L+ compared with those currently receiving 2L. However due to low sample size of TNBC patients, this observation should be taken with caution. Global health status was similar for EU3 and USA patients currently on 1L and 2L in both patient populations. For EU3 and US patients, EORTC scores showed a general trend for functioning and symptom scores being worse in patients receiving 2L and 3L+ compared with patients at 1L, in both patient populations; comparing patients at 1L and 2L, meaningful minimal differences were observed in global health status, cognitive functioning, dyspnea, and insomnia, as well as greater financial difficulties in both patient populations.

Patient-completed measurements of QoL are modified by psychological factors such as health expectancy, and therefore may change over time as the individual adjusts to living with the disease [32]. Thus, patients' assessment of QoL and the treatment effect may have increased or decreased at different therapy lines as they accommodated for disease progression and treatment effects over time. Nevertheless, many factors can affect and significantly impact patients' EORTC-QoL, such as age, disease stage at presentation, performance status, disease status at last follow-up, comorbidity, and remaining survival duration [27,33]. Symptoms may also influence QoL, with presence/severity of pain that frequently occurs due the disease or as a treatment side effect, being shown in previous studies to compromise the QoL of patients with breast cancer [28].

Global health status in our study was similar to that obtained in real-world studies reporting the global QoL of women treated for metastatic ABC [26,28]. While not examining different therapy lines, a review of eleven ABC studies observed that hormone therapy plus targeted therapy users experienced similar QoL in 1L relative to patients on hormone monotherapy [30]. Additionally, a study of women with metastatic ABC receiving palliative therapy in clinical practice found a significant relationship between therapy type and general quality of life, as measured by the EORTC-QLQC30 and QLQ-BR23 questionnaires; chemotherapy had no effect while hormone therapy and targeted therapy improved QoL [26].

Overall, physicians and patients appeared less satisfied with treatment at 2L and 3L+ compared with initial treatment, which was likely influenced by lower response rates for patients on later lines of therapy. As each line failed, it seems that patients lose confidence in the effectiveness of their therapy. Differences between the EU3 and USA by line of therapy were potentially driven by formulary and treatment guideline differences within the individual regions. Other studies have investigated treatment satisfaction in the USA and EU3. A survey in the USA found that better QoL, greater treatment satisfaction and better feelings about side effects were reported by patients with metastatic breast cancer using hormone therapy for one or more years than patients receiving chemotherapy [34]. However, a real-world study in Europe observed that while patients on chemotherapy were less satisfied with treatment and felt worse about side effects than patients on hormone therapy, therapy expectations did not differ by therapy type [35]. These findings are perhaps not unexpected considering hormone therapy and targeted therapy improved QoL while chemotherapy did not [26].

In interpreting this study's findings, its limitations should be considered. The DSP^™ was not based on a true random sample of physicians or patients. Despite minimal inclusion criteria, physician participation was influenced by their willingness to complete the survey. Physicians were asked to provide data for a consecutive series of patients to avoid selection bias, but no formal patient selection verification procedures were in place. This analysis likely represents the treatment satisfaction and QoL of a select patient population. Patients participating in the survey may not reflect the general ABC population; data were likely to be collected from patients who visit their physician more frequently. In addition, as the survey sampled patients at the point of their most recent physician consultation for routine care, data for patients who had previously passed away were not captured. Therefore, certain outcomes such as PFS, ORR as well as patient-reported outcomes relating to QoL and satisfaction with treatment could be overstated, particularly in patients at later stage of disease or at later lines of therapy. The low sample size of USA patients at 3L+ also limited interpretation and analysis of patient satisfaction and QoL data at this therapy line.

Additionally, all data collected relied on the accurate reporting of physicians and patients. As with all surveys, recall bias might also have affected the responses of patients and physicians to the questionnaires. However, data for these analyses were collected at the time of each patient's appointment and this was expected to reduce the likelihood of recall bias. In addition, physicians had access to patient medical records for extraction of retrospective data. Due to the anonymized and aggregated nature of data collection and analysis, no source validation was possible. The cross-sectional design of this study prevents any conclusions about causal relationships, although identification of associations was possible. This study was descriptive and no statistical analysis was performed. However, we used a number of standardized and validated instruments to assess quality of life. These instruments are associated with minimal important differences that indicate whether any differences observed between groups can be considered clinically meaningful, and where appropriate we present our results in the context of these values.

Although real-world studies do have limitations, data from these studies can complement evidence from clinical trials, which often have strict inclusion criteria and do not always reflect what happens in clinical practice [36,37]. Real-world studies provide insights into various aspects of treatment and outcomes from the perspective of both clinicians and their patients, reflecting clinical practice at the time the study was conducted.

Conclusion

This article is among the few real-world studies which describes outcomes in ABC, including QoL and treatment satisfaction, with patients stratified by line of current treatment. The results demonstrated that patients with HR+/HER2- Stage IIIb–IV ABC currently receiving 2L and 3L+ reported a lower QoL compared with patients on 1L therapy. For patients with TNBC Stage IIIb-IV ABC, those on 2L reported a lower QoL compared with patients on 1L. Concurrently, overall, median time to progression was shorter and ORR and DCR were poorer when comparing 1L to 3L patients in both patient populations. Similarly, satisfaction with treatment was reduced for both physicians and patients, in patients receiving later therapy lines. It appears that with decreasing response to therapy and therapy line failure, patients lose confidence that their therapy will work to improve QoL. Formularies and treatment guidelines within individual regions likely play a role in differences in treatment satisfaction between the EU3 and USA. While survival outcomes have improved with the introduction of new treatments, QoL and treatment satisfaction across therapy lines needs further investigation to improve the wellbeing of patients with ABC.

Author contributions

K Ndirangu was responsible for clinical oversight and guidance as lead author. All authors were involved in conception or design, or analysis and interpretation of data; drafting and revising the article; providing intellectual content of critical importance to the work described; and final approval of the version to be published, and therefore meet the criteria for authorship in accordance with the International Committee of Medical Journal Editors guidelines. In addition, all named authors take responsibility for the integrity of the work as a whole and have given their approval for this version to be published.

Financial disclosure

Eisai Inc. did not influence the original survey through either contribution to the design of questionnaires or data collection. The analysis described here used data from the Adelphi Real World Advanced Breast Cancer DSP. The DSP is a wholly owned Adelphi Real World product. Eisai Inc. is one of multiple subscribers to the DSP. Publication of survey results was not contingent on the subscriber's approval or censorship of the manuscript. Q Zhao and K Ndirangu are employees of Eisai Inc. Nutley, United States, and M Lucero and G Meier are former employees of Eisai Inc. K Lewis and A Lambert are employees of Adelphi Real World. I Chabot works in the Faculty of Pharmacy, University of Montreal, Canada and was commissioned by Eisai Inc. to support the study development. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Writing disclosure

Medical writing and editorial assistance were provided by S Libretto of Sue Libretto Publications Consultant Ltd (Hertfordshire, UK) on behalf of Adelphi Real World, in accordance with Good Publication Practice (GPP3) guidelines. Additional medical writing and editorial support on behalf of Adelphi Real World were provided by G Sidgwick of Adelphi Real World, Bollington, UK under the guidance of the authors and in accordance with Good Publication Practice (GPP3) guidelines.

Ethical disclosure

The study was performed in accordance with relevant guidelines; ethics approval was obtained from Western Institutional Review Board, study protocol number AG8643. Both patients and physicians provided informed consent to participate in the survey. Data were collected in such a way that patients and physicians could not be identified directly, and no personally identifiable information was collected.

Data sharing statement

All data, in other words, methodology, materials, data and data analysis, that support the findings of this survey are the intellectual property of Adelphi Real World. All requests for access should be addressed directly to Katie Lewis at [email protected]. Katie Lewis is an employee of Adelphi Real World.

Supplementary data

To view the supplementary data that accompany this paper please visit the journal website at: www.futuremedicine.com/doi/suppl/10.2217/fon-2022-1228

Competing interests disclosure

The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

Eisai Inc. did not influence the original survey through either contribution to the design of questionnaires or data collection.