1148921 research-article20232023 TAM0010.1177/17588359221148921Therapeutic Advances in Medical OncologyA Tibau, MT Martínez Therapeutic Advances in Medical Oncology Original Research Quality of life with palbociclib plus Ther Adv Med Oncol 2023, Vol. 15: 1­–14 fulvestrant versus placebo plus DOI: 10.1177/ https://doi.org/10.1177/17588359221148921 https://doi.org/10.1177/17588359221148921 17588359221148921 fulvestrant in postmenopausal women © The Author(s), 2023. Article reuse guidelines: sagepub.com/journals- with endocrine-sensitive hormone permissions receptor-positive and HER2-negative advanced breast cancer: patient-reported Correspondence to: Ariadna Tibau Hospital de la Santa outcomes from the FLIPPER trial Creu i Sant Pau, Institut d’Investigació Biomèdica Sant Pau, and Departament de Medicina de la Universitat Autònoma Ariadna Tibau , M. Teresa Martínez, Manuel Ramos, Luis De La Cruz-Merino, de Barcelona, C/ Sant Quintí, 89, Barcelona Ana Santaballa, Miriam O’Connor, Noelia Martínez-Jañez, Fernando Moreno, 08041, Spain Isaura Fernández, Juan Antonio Virizuela, Jesús Alarcón, Juan de La Haba-Rodríguez, GEICAM Spanish Breast Pedro Sánchez-Rovira, Cinta Rosa Albacar, Coralia Bueno Muiño, Catherine Kelly, Cancer Group, Madrid, Spain. Maribel Casas, Susana Bezares, Libertad Rosell and Joan Albanell atibau@santpau.cat M. Teresa Martínez GEICAM Spanish Breast Abstract Cancer Group, Madrid, Spain Background: In the FLIPPER trial, palbociclib/fulvestrant significantly improved progression- Hospital Clínico free survival (PFS) compared with placebo/fulvestrant in postmenopausal women with HR+/ Universitario de Valencia, Instituto de Investigación HER2− advanced breast cancer (ABC). Sanitaria-INCLIVA, Objective: We assessed health-related quality of life (QoL) using patient-reported outcomes (PROs). Valencia, Spain Design and methods: In this phase II double-blinded study, PROs were assessed at baseline Manuel Ramos GEICAM Spanish Breast after every three cycles and at the end of the treatment using the European Organisation Cancer Group, Madrid, Spain for Research and Treatment of Cancer QLQ-C30 and QLQ-BR23. Time to deterioration (TTD) Centro Oncológico de in global health status (GHS)/QoL was defined as a decrease of ⩾10 points. Changes from Galicia, Coruña, Spain baseline (CFB) and TTD were analysed using linear mixed-effect and Cox regression models, Luis De La Cruz-Merino GEICAM Spanish Breast respectively. Cancer Group, Madrid, Results: Of the 189 randomised (1:1) patients, 178 (94%) completed ⩾1 post-baseline Spain Hospital Universitario assessment; 50% received ⩾22 cycles of study treatment, with a questionnaire compliance Virgen de la Macarena, >90%. Mean baseline scores were comparable between arms. GHS/QoL scores were Medicine Department- Universidad de Sevilla, maintained throughout the palbociclib/fulvestrant treatment. CFB showed significant Sevilla, Spain differences for GHS/QoL, appetite loss, constipation and systemic therapy side effect scores Ana Santaballa GEICAM Spanish Breast favouring placebo/fulvestrant. TTD in GHS/QoL was delayed in placebo/fulvestrant versus Cancer Group, Madrid, palbociclib/fulvestrant [30.3 versus 11.1 months; adjusted hazard ratio (aHR): 1.57, 95% CI: Spain Hospital Universitario 1.03–2.39, p = 0.036]; this difference was not significant in patients with progressive disease y Politécnico La Fe, (aHR: 1.2, 95% CI: 0.6–2.2, p = 0.658). No statistically significant differences in TTD were found Valencia, Spain Miriam O’Connor for the other QLQ-C30 and QLQ-BR23 scales. University Hospital Conclusions: Although TTD in GHS/QoL was prolonged with placebo/fulvestrant, no Waterford, Waterford, Ireland differences were observed on other functional or symptom scales. This finding and the Cancer Trials Ireland, improvement in PFS support the combination of palbociclib/fulvestrant as a beneficial Dublin, Ireland therapeutic option for HR+/HER2− ABC. Noelia Martínez-Jañez GEICAM Spanish Breast Cancer Group, Madrid, Spain Trial registration number: Sponsor Study Code: GEICAM/2014-12 Hospital Universitario EudraCT Number: 2015-002437-21 Ramón y Cajal, Madrid, ClinTrials.gov reference: NCT02690480 Spain journals.sagepub.com/home/tam 1 Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). Therapeutic Advances in Medical Oncology Volume 15 Fernando Moreno Keywords: advanced breast cancer, CDK4/6 inhibitor, fulvestrant, palbociclib, patient- GEICAM Spanish Breast Cancer Group, Madrid, reported outcomes, quality of life Spain Hospital Clínico Received: 9 June 2022; revised manuscript accepted: 15 December 2022. Universitario San Carlos, Madrid, Spain Isaura Fernández GEICAM Spanish Breast Cancer Group, Madrid, Spain Introduction fulvestrant versus placebo/fulvestrant in postmen- Hospital Álvaro Cunqueiro, Vigo, Spain In the FLIPPER trial (ClinTrials.gov number, opausal women with HR+/HER2− ABC. Juan Antonio Virizuela NCT02690480), the efficacy and safety of the Patients had either de novo metastatic disease or GEICAM Spanish Breast CDK4/6 inhibitor palbociclib plus fulvestrant remained disease free for >12 months after com- Cancer Group, Madrid, Spain was compared with placebo plus fulvestrant for pleting at least 5 years of adjuvant ET. Patients Hospital QuirónSalud the treatment of postmenopausal woman with were assigned in a 1:1 ratio to receive palbociclib Sagrado Corazón, Sevilla, endocrine-sensitive hormone receptor-positive (125 mg/day, 28-day cycle; 3 weeks on, 1 week Spain (HR+) and human epidermal growth factor off) or matched placebo, and both arms received Jesús Alarcón GEICAM Spanish Breast receptor 2-negative (HER2−) advanced breast fulvestrant (500 mg on day 1 of each 28-day cycle, Cancer Group, Madrid, cancer (ABC). Primary results demonstrated a with an additional dose on day 15 of cycle 1). Spain Hospital Universitari statistically significant improvement for a number Treatment continued until objective progressive Son Espases, Palma de of metrics with palbociclib/fulvestrant as com- disease (PD) according to the Response Mallorca, Spain pared to placebo/fulvestrant. Specifically, 1-year Evaluation Criteria in Solid Tumours version Juan de La Haba- Rodríguez progression-free survival (PFS) rate [83.5% ver- 1.1,8 symptomatic deterioration, unacceptable GEICAM Spanish Breast sus 71.9%; hazard ratio (HR): 0.55, 80% confi- toxicity, death or withdrawal of consent, accord- Cancer Group, Madrid, Spain dence interval (CI): 0.36–0.83, p = 0.064], PFS ing to which occurred first. The detailed study Instituto Maimonides de [median 31.8 versus 22.0 months; adjusted HR design and characteristics of patients have been Investigacion Biomedica, (aHR): 0.48; 80% CI: 0.37–0.64, p = 0.001)] and reported previously.1 Hospital Reina Sofia Hospital, Universidad de overall response rate (68.3% versus 42.2%; odds Córdoba, Spain ratio: 2.9; 80% CI: 1.8–4.6, p = 0.004) with a Centro de Investigación manageable safety profile. Results on overall sur- PRO assessments Biomédica en Red de Oncología, CIBERONC- vival (OS) are still immature1 The comparison of the health-related QoL ISCIII, Madrid, Spain (HRQoL) between treatment arms as reported by Pedro Sánchez-Rovira GEICAM Spanish Breast The addition of CDK4/6 inhibitors to endocrine patients was a secondary objective of the trial Cancer Group, Madrid, therapy (ET) as first-line therapy for postmeno- using as secondary endpoints the changes from Spain pausal woman with HR+/HER2− ABC leads to a baseline (CFB) and time to deterioration (TTD) Hospital Universitario de Jaén, Jaén, Spain statistically and clinically significant improvement PRO measures of HRQoL were assessed using Cinta Rosa Albacar in PFS when compared with endocrine mono- the European Organisation for Research and GEICAM Spanish Breast therapy.2–4 However, combination treatments can Treatment of Cancer (EORTC) QoL question- Cancer Group, Madrid, Spain expose patients to additional treatment-related naire-Core 30 (QLQ-C30) V3.09 and its breast Hospital Universitario Sant toxicities, which may negatively affect their qual- cancer-specific module (QLQ-BR23; v1.0).10 Joan de Reus, Spain ity of life (QoL). Recommendations from medical Coralia Bueno Muiño GEICAM Spanish Breast oncology societies experts and regulators high- Patients were asked to complete each question- Cancer Group, Madrid, light the need for integration of patient-reported naire at baseline, every three cycles until the end Spain outcomes (PROs) into the discussion of efficacy of treatment, and at the post-treatment visit (per- Hospital Infanta Cristina (Parla), Fundación de and safety drug profile to better define their clini- formed 30 days after the last study treatment Investigación Biomédica cal benefit.5–7 Here, we report validated cancer- dose). Questionnaires were completed by patients del H.U, Puerta de Hierro Majadahonda, Madrid, related and breast cancer-specific PRO results in the clinic environment prior to any testing or Spain from the FLIPPER trial. discussion with healthcare personnel at the site. Catherine Kelly Cancer Trials Ireland, Dublin, Ireland Mater The EORTC QLQ-C30 is a 30-item question- Misericordiae University Material and methods naire composed of a global health status (GHS)/ Hospital, Ireland QoL multi-item scale, five multi-item functional Maribel Casas Susana Bezares Study design and patients scales (physical, role, emotional, cognitive and Libertad Rosell The FLIPPER is a phase II, randomised, interna- social), three multi-item symptom scales (fatigue, GEICAM Spanish Breast Cancer Group, Madrid, tional, multicentre, double-blinded, placebo-con- nausea/vomiting and pain) and six single-item Spain trolled trial comparing the efficacy of palbociclib/ symptom scales (dyspnoea, insomnia, appetite 2 journals.sagepub.com/home/tam A Tibau, MT Martínez et al. loss, constipation, diarrhoea and financial diffi- Table 1. Item numbers and definition of the MID as CFB values by scale in culties)9 (Table 1). The EORTC QLQ-BR23 the EORTC QLQ-C30 and QLQ-BR23 instruments. questionnaire is a 23-item companion module Instruments, scales Item numberb MIDa consisting of two multi-item functional scales (body image and sexual functioning) and three EORTC QLQ-C30 multi-item symptom scales (systemic therapy side Functional scales effects, arm and breast symptoms) and single   Physical functioning 1–5 6 item functional and symptom scales covering sex- ual enjoyment, future perspective and upset due   Role functioning 6–7 8 to hair loss10 (Table 1).   Emotional functioning 21–24 4 Two seven-point Likert scales for GHS/QoL with   Cognitive functioning 20, 25 2 responses from ‘very poor’ to ‘excellent’ and four-   Social functioning 26, 27 7 point Likert scales to assess functioning and QoL symptoms with responses from ‘not at all’ to ‘very much’ were provided.   GHS/QoL 29, 30 10 Symptom scales Responses to all item measures were converted   Fatigue 10, 12, 18 6 into linear scales ranging from 0 to 100 using a standard scoring algorithm.11 For the GHS/QoL    Nausea and vomiting 14, 15 6 and functional scales, a higher score represents a   Pain 9, 19 4 better level of QoL/functioning and an increase from baseline indicates an improvement in QoL.   Dyspnoea 8 6 By contrast, for symptomatic scales, a higher   Insomnia 11 3 numerical score represents greater or worse symp-   Appetite loss 13 3 tom severity.   Constipation 16 6   Diarrhoea 17 6 Statistical analysis   Financial difficulties 28 3 PROs analyses were performed in cases with baseline and at least one post-baseline assessment EORTC QLQ-BR-23 (QoL population). Completion rates were sum- Functional scales marised by visit in the intention-to-treat (ITT) population; a questionnaire was considered   Body image 9–12 5 received if at least one question was answered.   Sexual functioning 14, 15 5 For partially completed multi-item scales, miss-   Sexual enjoyment 16 5 ing scores were equal to the average of the com- pleted items if at least half of the items of that   Future perspective 13 5 scale were answered, but were not included in the Symptom scales analysis if less than that were completed.    Systemic therapy side effects 1–4, 6, 7, 8 5 Descriptive statistics, including 95% CI for the   Breast symptoms 20–23 5 means of actual values and CFB were tabulated at   Arm symptoms 17,18,19 5 the scheduled time points for each scale of the    Upset by hair loss 5 5 EORTC QLQ-C30 and EORTC QLQ-BR23 questionnaires. The means and 95% CIs of CFB, Source: Cocks et al.12 and Osoba et al.13 aA deterioration event is an increase of ⩾MID from baseline for the symptom scales and a as well as comparisons between treatment arms decrease of ⩾MID from baseline for the functional scales and GHS/QoL. A clinically meaningful with their respective p-values, were analysed is a decrease of ⩾MID from baseline for the symptom scales and an increase of ⩾MID from baseline for the functional scales and GHS/QoL. using a linear mixed model. The model factors bThe GHS/QoL items were rated on a seven-point Likert scales with responses from ‘very poor’ were treatment arms, time points, treatment–time to ‘excellent’ and the other QLQ-C30 items and all QLQ-BR23 items were rated on a four-point Likert scale from ‘not at all’ to ‘very much’. interaction terms, and stratification criteria, and CFB, change from baseline; EORTC QLQ-BR-23, European organisation for research and the covariates were the baseline scores. A ran- treatment of cancer breast-specific questionnaire; EORTC QLQ-C30, European organisation for research and treatment of cancer core questionnaire; GHS/QoL, global health status/QoL; MID, dom-intercept only model with a first-order minimally important difference. autoregressive covariance structure was used. journals.sagepub.com/home/tam 3 Therapeutic Advances in Medical Oncology Volume 15 Joan Albanell Baseline scores were compared between treat- Table S1). Median age was 64 years, 97% of GEICAM Spanish Breast ment arms using t-test. patients had an Eastern Cooperative Oncology Cancer Group, Madrid, Spain Group performance status of zero or one, 45.5% Centro de Investigación TTD investigated both in the entire study QoL had de novo metastatic disease and among those Biomédica en Red de Oncología, CIBERONC- population and in subgroups, according to ther- having non-de novo metastatic disease, 39.7% had ISCIII, Madrid, Spain apy response (with/without PFS event), was a treatment-free interval after adjuvant ET greater Hospital del Mar Medical defined as the time from the date of randomisa- than 36 months. Most patients (60.3%) had vis- Research Institute (IMIM), Hospital del Mar, tion to the date of first detection of a deterioration ceral involvement; overall, 33.9% patients had Barcelona, Spain event. The minimally important difference (MID) lung involvement, 15.9% liver involvement and Pompeu Fabra University, is the change in score of a PRO that is important 15.9% bone-only disease. Barcelona, Spain from the patient’s or clinician’s perspective and would warrant a change in the patient’s manage- ment. The MID values used were between 2 and Completion rates 10 points according to data previously pub- All PRO analyses were based on a cut-off date of lished.12,13 A deterioration event was defined as 11 January 2020, at a median follow-up time of an increase of ⩾MID from baseline for the 28.6 months (range, from 1.5 to 44.8). At this EORTC QLQ-C30 and QLQ-BR23 symptom point, 50% of patients had received the first 22 scales and a decrease of ⩾MID from baseline for cycles of study treatment and the overall ques- the EORTC QLQ-C30 and QLQ-BR23 func- tionnaires compliance was high in both treatment tional and GHS scales. Results were defined as arms, with a completion rate ⩾96% at baseline, clinically meaningful based on MIDs for EORTC ⩾92% on treatment and ⩾84% at post-treatment QLQ-C30 and QLQ-BR23 (Table 1). Patients visit (Supplemental Table S2). with no definitive deterioration event were cen- sored at their last available QoL assessment. In The compliance rates of each EORTC QLQ- patients with no post-baseline assessment, TTD C30, EORTC QLQ-BR23 scale were like the was censored on day 1. global questionnaire completion rates, except for sexual enjoyment and, upset due to hair loss, The Kaplan–Meier method was used to estimate where sample sizes were smaller versus other the distribution of TTD for each treatment arm scales (Table 2). and according to PFS event. A log-rank test was performed to compare the TTD between treat- ment arms. HRs were adjusted by stratification EORTC QLQ-C30 and QLQ-BR23 baseline scores factors: disease site (visceral versus non-visceral) Mean baseline scores were similar in all dimen- and (recurrent versus de novo metastatic disease). sions among the treatment arms, except for the Adjusted HR (aHRs) and two-sided 95% CIs body image, with better level of functioning in were estimated using Cox regression model for placebo/fulvestrant arm (Table 2). Globally, all the comparison of treatment arms. baseline scores were in line with the EORTC QLQ-C30, QLQ-BR23 reference values expected for ABC.14 Results Patient characteristics CFB per treatment arm Between February 2016 and January 2018, 189 Palbociclib/fulvestrant. The combination of pal- patients were recruited across 32 institutions in bociclib/fulvestrant showed a GHS/QoL mean two countries [177 patients in Spain (GEICAM) scores maintained numerically up to cycle 22, and and 12 in Ireland (Clinical Trials Ireland)]. The a decrease occurred at the post-treatment visit QoL population comprised 178 (94.2%) patients (Figure 1). When analysing the mean CFB by with completed baseline and ⩾1 post-baseline visit, significant improvement at different time PROs. Of these, 88 were included in the palboci- points was found for emotional functioning, social clib/fulvestrant arm and 90 in the placebo/fulves- functioning, body image, pain, dyspnoea, diar- trant arm (Supplemental Figure S1). rhoea, breast symptom and arm symptom (Sup- plemental Table S3). Meanwhile, significant Baseline demographic and disease characteristics impairment was found for future perspective, were well balanced between arms (Supplemental constipation, financial difficulties and systemic 4 journals.sagepub.com/home/tam A Tibau, MT Martínez et al. Table 2. Baseline mean scores of EORTC QLQ-C30 and EORTC QLQ-BR23 scales by treatment arm. Palbociclib/fulvestrant n = 88 Placebo/fulvestrant n = 90 p Value t-test Reference valuesa n Mean (SD) n Mean (SD) EORTC QLQ-C30 functional scales Physical functioning 88 75.4 (21.3) 90 77.9 (23.1) 0.445 81.6 (18.7) Role functioning 88 72.2 (29.4) 90 72.2 (33.2) 0.989 67.4 (31.1) Emotional functioning 88 63.5 (25.8) 90 69.2 (23.5) 0.121 65.9 (24.6) Cognitive functioning 88 87.5 (16.8) 90 88.3 (19.7) 0.762 80.5 (23.2) Social functioning 87 79.1 (25.3) 88 85.2 (21.7) 0.088 74.2 (28.4) EORTC QLQ-C30 QoL GHS/QoL 88 60.0 (24.2) 90 57.0 (25.6) 0.423 60.2 (25.5) EORTC QLQ-C30 symptom scales Fatigue 88 34.0 (26.0) 90 29.3 (27.1) 0.238 36.3 (27.0) Nausea and vomiting 88 2.8 ( 6.8) 90 5.4 (14.7) 0.143 10.3 (19.7) Pain 88 35.6 (31.7) 90 30.6 (30.2) 0.278 30.9 (29.6) Dyspnoea 88 22.7 (31.4) 90 15.6 (25.1) 0.094 20.4 (28.2) Insomnia 88 26.9 (30.3) 89 33.3 (31.4) 0.167 33.1 (32.6) Appetite loss 88 19.3 (26.6) 90 19.3 (29.1) 0.989 21.7 (31.0) Constipation 88 20.1 (30.1) 90 17.8 (26.1) 0.587 19.2 (28.8) Diarrhoea 88 6.8 (14.4) 90 5.2 (13.1) 0.431 5.8 (15.2) Financial difficulties 87 6.1 (14.9) 89 12.7 (26.8) 0.045 18.6 (28.6) EORTC QLQ-BR23 functional scales Body image 88 79.4 (29.5) 90 88.7 (22.6) 0.019 81.9 (22.6) Sexual functioning 79 12.0 (20.3) 86 11.6 (20.3) 0.900 19.2 (23.2) Sexual enjoymentb 30 33.3 (32.8) 26 41.0 (36.9) 0.412 55.1 (25.6) Future perspective 86 41.1 (32.6) 87 43.3 (36.0) 0.673 47.6 (34.1) EORTC QLQ-BR23 symptom scales Systemic therapy side effects 87 18.0 (16.0) 89 15.9 (12.7) 0.339 15.8 (14.3) Breast symptoms 88 20.4 (23.4) 90 15.2 (19.3) 0.110 17.6 (16.7) Arm symptoms 88 21.5 (28.5) 90 15.5 (20.7) 0.115 21.0 (21.1) Upset by hair lossc 22 15.2 (24.6) 24 9.7 (15.5) 0.382 5.3 (19.3) aReference baseline value for recurrent and metastatic breast cancer patients across all lines of treatment.12 bThe patients were asked to answer this question only if they responded that they were sexually active in a previous question. cOnly patients with alopecia were asked to answer this question. EORTC QLQ-BR-23, European organisation for research and treatment of cancer breast-specific questionnaire; EORTC QLQ-C30, European organisation for research and treatment of cancer core questionnaire; QoL, quality of life; SD, standard deviation. journals.sagepub.com/home/tam 5 Therapeutic Advances in Medical Oncology Volume 15 Comparison of CFB between treatment arms The overall CFB for all scales of the QLQ-C30 and QLQ-BR23 questionnaires are presented in Figure 3. Significant improvements were observed for GHS/QoL [6.8 (95% CI: 3.5–10.0) versus 0.2 (95% CI: −3.2–3.6); p = 0.005], and appetite loss [−7.3 (95% CI: −10.4–4.3) versus −0.6 (95% CI: −5.7–4.5); p = 0.021], and significantly less dete- rioration for constipation [2.5 (95% CI: −2.0– 6.9) versus 10.5 (95% CI: 5.0–15.9); p = 0.027] and for systematic therapy side effects [2.7 (95% CI: 0.2–5.1) versus 8.3 (95% CI: 5.2–11.4); p = 0.004] favouring the control arm with pla- cebo/fulvestrant. Figure 1. GHS/QoL mean scores in the EORTC QLQ-C30 scales by visit. BL, baseline; C, cycle; CI, confidence interval; D, day; EORTC QLQ-C30, European According to the linear mixed model analysis, the organisation for research and treatment of cancer core questionnaire; FUL, change of GHS/QoL score from baseline to cycle fulvestrant; GHS/QoL, global health status/quality of life; PAL, palbociclib; PBO, placebo; POST-T, post-treatment visit. 7 was 1.5 (95% CI: −2.5–5.5) for palbociclib/ful- vestrant versus 7.9 (95% CI: 3.4–12.4) for pla- cebo/fulvestrant. The mean difference was −6.4 (95% CI: −12.4–0.5; p = 0.035), maintained up therapy side effects when the analysis of mean to cycle 19, showing a higher improvement in the CFB by visit was performed (Supplemental Table placebo/fulvestrant arm. No significant differ- S4). No significant differences were observed for ences were observed between treatment arms at other functional, symptom or GHS/QoL scales. other time points (Supplemental Figure S2). Placebo/fulvestrant. The GHS/QoL mean scores were numerically improved up to post-treatment Time to deterioration visit (Figure 1). Statistically significant increases The median TTD in GHS/QoL, using a in the mean CFB during all study treatment visits MID = 10, was 11.1 months with palbociclib/ful- were found for GHS/QoL (Table S3). vestrant versus 30.3 months with placebo/fulves- trant (aHR: 1.57, 95% CI: 1.03–2.39; p = 0.036) Significant improvements were found at different ((Figure 4(a)). The analysis performed only in time points in the mean CFB by visit for role func- patients without a PFS event showed similar tioning, emotional functioning and appetite loss results, 11.1 months with palbociclib/fulvestrant (Table S3). Significant impairment was found for versus 30.3 months with placebo/fulvestrant (aHR cognitive functioning, future perspective and sys- 2.04: 95% CI: 1.11–3.77; p = 0.023). No statisti- temic therapy side effects (Supplemental Table cally significant difference was seen in TTD in S4). Most scales worsened in both study arms at the analysis performed in patients with a PFS the post-treatment visit since the reason for termi- event (aHR: 1.15, 95% CI: 0.61–2.18; p = 0.658) nating the treatment was PD in 82.4% of the cases (Figures 4(b) and (c)). (Supplemental Table S5). No significant differences were seen in that respect among patients with or without a PFS Clinically meaningful improvements event in both arms, palbociclib/fulvestrant (aHR: Clinically meaningful improvements (based on 1.21; 95% CI: 0.69–2.11, p = 0.504) and placebo/ MID values) were observed in both treatment fulvestrant (aHR: 0.55; 95% CI: 0.26–1.13, arms at different time points for emotional func- p = 0.104) (Supplemental Figure S3). tioning, pain, appetite loss and breast symptoms; while such improvements developed for arm No significant differences were seen for the risk of symptoms in the palbociclib/fulvestrant arm, in deterioration between the study arms of the the placebo/fulvestrant arm improvements median TTD for the other QLQ-C30 and developed for role functioning and dyspnoea QLQ-BR23 scales, except for arm symptoms, (Figure 2). using a MID = 5, with a median TTD of 6 journals.sagepub.com/home/tam A Tibau, MT Martínez et al. Figure 2. Clinically meaningful improvements from baseline in the EORTC QLQ-C30 and QLQ-BR23 scales by treatment arm. Baseline is defined as the last observed measurement on or before the date of first dose of study drug. The time profile provides the average estimates for the CFB for the interval from baseline to the respective cycle as derived from the linear mixed model. An increase from baseline indicates an improvement in functional items (a – emotional functioning and f – role functioning]. A decrease from baseline indicates an improvement in symptom scales (b – pain, c – appetite loss, d – breast symptoms, e – arms symptoms and g – dyspnoea). MID values indicated by red lines (Cocks et al.12 and Osoba et al.13) have been used for the consideration of CMI. C, cycle; CFB, change from baseline; CI, confidence interval; CMI, clinically meaningful improvements; D, day; EORTC QLQ-BR-23, European organisation for research and treatment of cancer breast-specific questionnaire; EORTC QLQ-C30, European organisation for research and treatment of cancer core questionnaire; MID, minimally important difference; PAL, palbociclib; PBO, placebo; POST-T, post-treatment visit. journals.sagepub.com/home/tam 7 Therapeutic Advances in Medical Oncology Volume 15 Figure 3. Overall CFB in EORTC QLQ-C30 and BR23 scales. CFB were determined using a repeated-measures mixed-effect model. (a) Analysis of CFB for GHS/QoL and functional scales and (b) analysis of CFB for symptom scales. *Statistically significant difference in CFB scores between treatment arms. EORTC QLQ-BR-23, European organisation for research and treatment of cancer breast-specific questionnaire; EORTC QLQ-C30, European organisation for research and treatment of cancer core questionnaire; CFB, change from baseline; FUL, fulvestrant; GHS/QoL, global health status/quality of life; PAL, palbociclib; PBO, placebo. 20.4 months with palbociclib/fulvestrant versus fulvestrant administration in the first-line treat- 8.3 months with placebo/fulvestrant (aHR: 0.59, ment of endocrine-sensitive HR+/HER2− ABC 95% CI: 0.39–0.88; p = 0.011). In the Forest-plot patients. We demonstrated that the favourable with the QLQ-C30 and QLQ-BR23 scales for efficacy achieved with palbociclib/fulvestrant is TTD, a trend towards one of the treatment arms accompanied by maintained overall QoL cannot be observed, the distribution from the unit throughout the study treatment period. towards both sides is heterogenous (Figure 5). Interestingly, baseline GHS/QoL and function- ing subscale scores were maintained and similar Subgroup analyses of TTD in GHS/QoL showed between study arms, except for GHS/QoL, longer delayed deterioration with placebo/fulves- appetite loss, constipation and systemic therapy trant over palbociclib/fulvestrant (Supplemental side effects, which statistically favoured the con- Figure S4). trol over time. Appetite loss, constipation, and systemic therapy side effects, as treatment- related toxicities that can adversely affect the Discussion HRQoL, could have contributed to the reduc- Maintaining QoL is a key goal of the treatment tion in GHS/QoL observed in our study. Indeed, of ABC patients. Although treatments can this is a relative reduction compared to the pla- reduce symptoms, thereby slowing the impair- cebo arm and not an absolute reduction, since ment in QoL and prolonging time to PD, treat- the GHS/QoL scores in the experimental arm ment-related toxicities affect the HRQoL. Here, are maintained and numerically improved dur- we present detailed cancer-related and breast- ing treatment compared to baseline values. In cancer-specific PROs regarding the palbociclib/ addition, declines in these symptom scales were 8 journals.sagepub.com/home/tam A Tibau, MT Martínez et al. Figure 4. Kaplan–Meier estimates for TTD in GHS/QoL based on the EORTC QLQ-C30 questionnaire data. Baseline is defined as the last observed measurement on or before the date of first dose of study drug. The aHR was obtained using a stratified Cox proportional hazards model with treatment arm and the stratification factors [disease site (visceral versus non-visceral) and (recurrent versus de novo) metastatic disease] as covariates. (a) Analysis of all QoL population by treatment arm, (b) analysis of patients without progression event and (c) analysis of patients with progression event. CI, confidence interval; EORTC QLQ-C30, European organisation for research and treatment of cancer core questionnaire; FUL, fulvestrant; GHS/QoL, global health status/quality of life; PAL, palbociclib; PBO, placebo; TTD, time to deterioration. consistent with the safety profile observed with Although no statistical or clinically meaningful palbociclib/fulvestrant in the FLIPPER trial.1 In differences were observed in the key subdomains the ITT population, most adverse events (AEs) of the EORTC QLQ-C30 and QLQ-B23 ques- observed in the experimental arm were of mild tionnaires, patients treated with placebo/fulves- severity (except for neutropenia), with higher trant experienced a significantly greater delay in rates of Common Terminology Criteria for deterioration of GHS/QoL than patients treated Adverse Events grade I-II anorexia reported in with palbociclib/fulvestrant. In addition, patients the palbociclib/fulvestrant group (16%) versus treated with ET in monotherapy without PD the placebo/fulvestrant arm (9.5%) and no dif- experienced delayed TTD in GHS/QoL but not ferences in grade I-II constipation (19%). patients whose disease progressed, suggesting Permanent discontinuation of the whole study that negative impacts on functioning scores may treatment due to AEs was similar in both arms be primarily due to the study treatment AEs. and occurred in four patients (4.3%) receiving palbociclib/fulvestrant and 4 patients (4.2%) These PROs from the FLIPPER trial add to the receiving placebo/fulvestrant. Overall, these growing body of evidence reported of CDK4/6 results may reflect the importance of chronic inhibitors trials.16 For postmenopausal patients, low-grade toxicities in patient experience, and the majority of trials report that HRQOL has been how in regulatory clinical trials, these are gener- maintained in the first-line setting17–20 or improved ally not captured well.15 in the second-line or later-line setting.21 Our journals.sagepub.com/home/tam 9 Therapeutic Advances in Medical Oncology Volume 15 Figure 5. Forest plot TTD in the various scales of the EORTC QLQ-C30 and EORTC QLQ_BR23 questionnaires. The aHRs were obtained using a stratified Cox proportional hazards model with treatment arm, the stratification factors [disease site (visceral versus non-visceral) and (recurrent versus de novo) metastatic disease] as covariates. aHR, adjusted hazard ratio; CI, confidence interval; EORTC QLQ-BR-23, European organisation for research and treatment of cancer breast-specific questionnaire; EORTC QLQ-C30, European organisation for research and treatment of cancer core questionnaire; FUL, fulvestrant; GHS/QoL, global health status/quality of life; MID, minimally important difference; PAL, palbociclib; PBO, placebo. results are not fully consistent with those observed maintaining HRQOL in treatment-naïve post- in the PALOMA-3 trial, where premenopausal menopausal women with ER+/HER2− MBC.24 and postmenopausal patients with endocrine- However, in PALOMA-2 study, HRQOL deteri- resistant HR+/HER2− metastatic breast cancer oration was delayed in patients with response and/ were randomised to palbociclib/fulvestrant versus or no progression. This inconsistency regarding placebo/fulvestrant. This trial demonstrated HRQOL may be due to type of ET and less sensi- improved PFS and a non-statistically significant tivity to ET (22% of patients with a relapse increase in OS22 with a delay in the TTD of QoL ⩽12 months) into the PALOMA-2 trial.2 Finally, and pain symptoms.23 This inconsistency regard- when palbociclib in combination with ET has ing QoL may reflect a younger, more heavily pre- been compared versus chemotherapy (capecit- treated, population with fewer comorbidities abine) in a more heavily pre-treated, endocrine included into the PALOMA-3 trial. Conversely, resistance population HRQOL was maintained the FLIPPER trial recruited only postmenopausal regardless of menopausal setting,21,25 albeit in patients with ABC, older (median age was 64 in PEARL trial patients showed earlier HRQOL the FLIPPER versus 57 in the PALOMA-3 trial), deterioration with chemotherapy.21 sensitive to ET, less symptomatic and who were receiving the first-line therapy. Sensitivity was Our results shed light on benefit–risk assessments defined as de novo ABC or recurrent disease ⩾1 to inform decision-making for palbociclib in post- year after the completion of ⩾5 years of (neo) menopausal patients with ABC. Maintaining adjuvant ET, a patient population specifically HRQoL in patients with ABC is crucial, espe- excluded in the PALOMA-3 trial. Similar results cially for drugs that improve PFS with no OS were observed with palbociclib over placebo in benefit. Efficacy data recently published1 shown PALOMA-2 trial, where the addition of palboci- OS data were still immature at the time of this clib to letrozole significantly improved PFS while analysis with only a 15% of ITT population with 10 journals.sagepub.com/home/tam A Tibau, MT Martínez et al. an OS event. Ongoing follow-up will help for a Committee of the Cork Teaching Hospitals, better knowledge of the risk–benefit profile of pal- Cork, Ireland (no approval numbers available). bociclib (expected date of publication 2025). All patients gave informed consent in writing. Data were analysed by a statistician employed by Potential limitations of FLIPPER study are as fol- GEICAM. lows: (i) a small number of participants in a phase II study (ii) and the number of cycles included in Consent for publication the analysis. The cut-off of this analysis, not pre- Not applicable. planned, was cycle 22 because at this point 50% of patients had received the study treatment and Author contribution(s) the overall questionnaires compliance was high in Ariadna Tibau: Investigation; Resources; both treatment arms. Therefore, delayed TTD Supervision; Writing – original draft; Writing – were not captured and ⩾50% of patients were review & editing. still on treatment at the analysis cut-off date. M. Teresa Martínez: Data curation; Inves­ Further follow-up and updated data collection tigation; Resources; Visualisation; Writing – will help to mitigate this bias. Another potential review & editing. limitation in the design of PRO analyses should be considered as data cannot be assumed missing Manuel Ramos: Data curation; Investigation; at random because frail patients are likely to have Resources; Visualisation; Writing – review & lower compliance and bias could occur as a result. editing. To address this limitation, the mixed-model Luis de la Cruz-Merino: Data curation; approach was chosen as well as the high compli- Investigation; Resources; Visualisation; Writing ance rates of treated patients in all study visits – review & editing. minimised the impact of this limitation. Despite these, strength of FLIPPER study are (i) a ran- Ana Santaballa: Data curation; Investigation; domised double-blinded design, (ii) high compli- Resources; Visualisation; Writing – review & ance rates ⩾90% and (iii) the study included editing. multiple instruments with a dedicated question- Miriam O’Connor: Data curation; Inves­ naire to identify breast-cancer-specific functions tigation; Resources; Visualisation; Writing – and symptoms and differentiated MID values for review & editing. each multi-item scale of EORTC QLQ-C30 and EORTC QLQ-BR23 modules. Noelia Martínez-Jañez: Data curation; Inves­ tigation; Resources; Visualisation; Writing – review & editing. Conclusions Fernando Moreno: Data curation; Investigation; Our findings provide evidence that overall Resources; Visualisation; Writing – review & HRQoL in the FLIPPER trial was maintained in editing. postmenopausal women with endocrine-sensitive HR+/HER2− ABC receiving palbociclib in com- Isaura Fernández: Data curation; Investigation; bination with fulvestrant as first-line treatment. Resources; Visualisation; Writing – review & Taken together, the improved clinical efficacy, editing. manageable safety profile and PRO results pro- Juan Antonio Virizuela: Data curation; vide a meaningful assessment of the benefits, risks Investigation; Resources; Visualisation; Writing and tolerability of palbociclib/fulvestrant in this – review & editing. patient population. Jesús Alarcón: Data curation; Investigation; Resources; Visualisation; Writing – review & Declarations editing. Juan de La Haba-Rodríguez: Data curation; Ethics approval and consent to participate Investigation; Resources; Visualisation; Writing The study protocol was approved by every site’s – review & editing. institutional review board (IRB) and every national regulatory agency. The IRBs that Pedro Sánchez-Rovira: Data curation; Inves­ approved the study were Clínica Parc de Salut tigation; Resources; Visualisation; Writing – Mar, Barcelona, Spain; Clinical Research Ethics review & editing. journals.sagepub.com/home/tam 11 Therapeutic Advances in Medical Oncology Volume 15 Cinta Rosa Albacar: Data curation; Inves­ funding from MSD-Merck, Roche Farma, Celgene; tigation; Resources; Visualisation; Writing – speaking fees from MSD-Merck, Roche Farma, review & editing. Bristol-Myers-Squibb and Amgen; grant support from Bristol-Myers-Squibb and Roche Farma. Dr. Coralia Bueno Muiño: Data curation; Santaballa has received consulting or advisory role Investigation; Resources; Visualisation; Writing fees from GSK, Clovis, MSD, AstraZeneca, Roche – review & editing. and Pfizer; speakers’ honoraria from GSK, Clovis, Catherine Kelly: Data curation; Methodology; Roche, MSD, AstraZeneca and Pfizer; and grant Resources; Visualisation; Writing – review & support by Pfizer, GSK and MSD. Dr. Martínez- editing. Jañez has received advisory board fees from Roche Farma, Astra Zeneca, Pfizer, Novartis, Lilly, Maribel Casas: Formal analysis; Visualisation; Daiichi Sankyo, Agendia and Eisai. Dr. Moreno Writing – original draft; Writing – review & has received advisory board fees from Roche/ editing. Genentech, Novartis, Pfizer, AstraZeneca, MSD- Susana Bezares: Methodology; Project admin- Merck and Daiichi Sankyo/AstraZeneca; speaker’s istration; Supervision; Visualisation; Writing – bureau fees from Pfizer; travel accommodations original draft; Writing – review & editing. expenses from Roche/Genentech, Pfizer, Novartis, Daiichi Sankyo/Astra Zeneca. Dr. Fernández has Libertad Rosell: Project administration; received consultant or advisory fees from MSD- Visualisation; Writing – review & editing. Astra Zeneca; speaker’s bureau fees from Pfizer, Joan Albanell: Conceptualisation; Data cura- AstraZeneca, Roche, MSD, Clovis and GSK. Dr. tion; Funding acquisition; Investigation; Alarcón: has received honoraria from GSC, Clovis, Methodology; Project administration; Resources; Roche and AstraZeneca; consulting or advisory Supervision; Visualisation; Writing – original board honoraria from GSK and Clovis; speaker draft; Writing – review & editing. and expert testimony fees from GSK, Clovis and Roche; and travel and accommodation support Acknowledgements from GSK. Dr. de La Haba-Rodríguez has received We acknowledge the patients who participated in speaker’s honoraria from AstraZeneca, Pfizer, the FLIPPER study and their families and all the Novartis and Lilly. Dr. Sánchez-Rovira has participating investigators. We also acknowledge received honoraria from Novartis, Lilly, Pfizer, the support of the staff at each study site and at Roche and Seagen. Dr. Bueno Muiño has received both GEICAM and Cancer Trials Ireland head- speaker fees from Novartis, Lilly, AstraZeneca, quarters. We would like to thank Elsevier BMS, Pfizer and GSK. Dr. Albanell has received Language Editing Services for English language consulting or advisory role fees from Roche, Pfizer, editing. Amgen, MSD, Lilly and Daiichi-Sankyo; research funding or grant support trials by Roche, Pfizer, Funding Amgen, MSD, Lilly, Daiichi-Sankyo; and travel The authors disclosed receipt of the following and accommodation support from Roche, Pfizer, financial support for the research, authorship Amgen, MSD, Lilly and Daiichi-Sankyo. All and/or publication of this article: The trial spon- remaining authors have declared no conflicts of sor is GEICAM Spanish Breast Cancer Group. interest. Two funding companies provided the following: GEICAM has received research funding from AstraZeneca provided fulvestrant and study Roche/Genentech, Bristol-Myers Squibb, grant, and Pfizer Inc. provided palbociclib and Novartis, Pfizer, Celgene, AstraZeneca, Merck placebo. Sharp & Dohme, Pierre Fabre and Takeda. Competing interests A complete list of the FLIPPER trial collabora- Dr. Tibau has received speaker fees from Novartis tors is provided in the Supplemental Appendix. and Eisai. Dr. Ramos has received speaker’s bureau from Novartis, Pfizer and Roche Farma. Dr. de La Availability of data and materials Cruz-Merino has received advisory board fees from The datasets used and/or analysed during this MSD-Merk, Roche Farma, Bristol-Myers-Squibb, study are available from the corresponding author Pierre-Fabré, Amgen and Novartis; research on reasonable request. 12 journals.sagepub.com/home/tam A Tibau, MT Martínez et al. ORCID iD 11. Fayers PM. Interpreting quality of life data: Ariadna Tibau https://orcid.org/0000-0003- population-based reference data for the EORTC 0229-6987 QLQ-C30. Eur J Cancer 2001; 37: 1331–1334. 12. Cocks K, King MT, Velikova G, et al. Evidence- Supplemental material based guidelines for interpreting change scores for Supplemental material for this article is available the European organisation for the research and online. treatment of cancer quality of life questionnaire core 30. Eur J Cancer 2012; 48: 1713–1721. 13. Osoba D, Rodrigues G, Myles J, et al. Interpreting the significance of changes in health- References related quality-of-life scores. J Clin Oncol 1998; 1. Albanell J, Martinez MT, Ramos M, et al. 16: 139–144. Randomized phase II study of fulvestrant plus palbociclib or placebo in endocrine-sensitive, 14. Scott NW, Fayers PM, Aaronson NK, Bottomley hormone receptor-positive/HER2-advanced A, de Graeff A, Groenvold M, Gundy C, Koller breast cancer: GEICAM/2014-12 (FLIPPER). M, Petersen MA and Mirjam AG Sprangers Eur J Cancer 2022; 161: 26–37. on behalf of the EORTC Quality of Life Group. EORTC QLQ-C30 reference values 2. Finn RS, Martin M, Rugo HS, et al. Palbociclib 2008. https://www.eortc.org/app/uploads/ and letrozole in advanced breast cancer. N Engl J sites/2/2018/02/reference_values_manual2008.pdf Med 2016; 375: 1925–1936. (2008, accessed 10 January 2023). 3. Im SA, Lu YS, Bardia A, et al. Overall survival 15. Hadji P, Jackisch C, Bolten W, et al. with ribociclib plus endocrine therapy in breast COMPliance and arthralgia in clinical therapy: cancer. N Engl J Med 2019; 381: 307–316. the COMPACT trial, assessing the incidence of 4. Goetz MP, Toi M, Campone M, et al. arthralgia, and compliance within the first year of MONARCH 3: abemaciclib as initial therapy for adjuvant anastrozole therapy. Ann Oncol 2014; advanced breast cancer. J Clin Oncol 2017; 35: 25: 372–377. 3638–3646. 16. Ribnikar D, Volovat SR and Cardoso F. 5. Cherny NI, Dafni U, Bogaerts J, et al. ESMO- Targeting CDK4/6 pathways and beyond in magnitude of clinical benefit scale version 1.1. breast cancer. Breast 2019; 43: 8–17. Ann Oncol 2017; 28: 2340–2366. 17. Goetz MP, Martin M, Tokunaga E, et al. 6. Schnipper LE, Davidson NE, Wollins DS, Health-related quality of life in MONARCH 3: et al. Updating the American society of clinical abemaciclib plus an aromatase inhibitor as initial oncology value framework: revisions and therapy in HR+, HER2- advanced breast cancer. reflections in response to comments received. Oncologist 2020; 25: e1346–e1354. J Clin Oncol 2016; 34: 2925–2934. 18. Verma S, O’Shaughnessy J, Burris HA, et al. Health- 7. Cardoso F, Paluch-Shimon S, Senkus E, et al. 5th related quality of life of postmenopausal women ESO-ESMO international consensus guidelines with hormone receptor-positive, human epidermal for advanced breast cancer (ABC 5). Ann Oncol growth factor receptor 2-negative advanced breast 2020; 31: 1623–1649. cancer treated with ribociclib + letrozole: results from MONALEESA-2. Breast Cancer Res Treat 8. Eisenhauer EA, Therasse P, Bogaerts J, et al. 2018; 170: 535–545. New response evaluation criteria in solid tumours: revised RECIST guideline (version 19. Fasching PA, Beck JT, Chan A, et al. Ribociclib 1.1). Eur J Cancer 2009; 45: 228–247. plus fulvestrant for advanced breast cancer: Health-related quality-of-life analyses from the 9. Aaronson NK, Ahmedzai S, Bergman B, et al. MONALEESA-3 study. Breast 2020; 54: 148–54. The European organization for research and treatment of cancer QLQ-C30: a quality-of-life 20. Cardoso F, Senkus E, Costa A, et al. 4th ESO- instrument for use in international clinical trials in ESMO international consensus guidelines for oncology. J Natl Cancer Inst 1993; 85: 365–376. advanced breast cancer (ABC 4)dagger. Ann Oncol 2018; 29: 1634–1657. 10. Sprangers MA, Groenvold M, Arraras JI, et al. The European organization for research and 21. Kahan Z, Gil-Gil M, Ruiz-Borrego M, et al. treatment of cancer breast cancer-specific quality- Health-related quality of life with palbociclib of-life questionnaire module: first results from a plus endocrine therapy versus capecitabine three-country field study. J Clin Oncol 1996; 14: in postmenopausal patients with hormone 2756–2768. receptor-positive metastatic breast cancer: journals.sagepub.com/home/tam 13 Therapeutic Advances in Medical Oncology Volume 15 patient-reported outcomes in the PEARL study. 24. Rugo HS, Dieras V, Gelmon KA, et al. Impact Eur J Cancer 2021; 156: 70–82. of palbociclib plus letrozole on patient-reported health-related quality of life: results from 22. Turner NC, Slamon DJ, Ro J, et al. Overall the PALOMA-2 trial. Ann Oncol 2018; 29: survival with palbociclib and fulvestrant in 888–894. advanced breast cancer. N Engl J Med 2018; 379: 1926–1936. 25. Lee S, Im SA, Kim GM, et al. Patient-reported 23. Harbeck N, Iyer S, Turner N, et al. Quality outcomes of palbociclib plus exemestane of life with palbociclib plus fulvestrant in with GnRH agonist versus capecitabine in Visit SAGE journals online previously treated hormone receptor-positive, premenopausal women with hormone receptor- journals.sagepub.com/ HER2-negative metastatic breast cancer: patient- positive metastatic breast cancer: a prospective, home/tam reported outcomes from the PALOMA-3 trial. open-label, randomized phase ll trial (KCSG-BR SAGE journals Ann Oncol 2016; 27: 1047–1054. 15-10). Cancers (Basel) 2020; 12: 3265. 14 journals.sagepub.com/home/tam