Advertisement

Patient-reported outcomes in patients with a germline BRCA mutation and HER2-negative metastatic breast cancer receiving olaparib versus chemotherapy in the OlympiAD trial

Open AccessPublished:August 22, 2019DOI:https://doi.org/10.1016/j.ejca.2019.06.023

      Highlights

      • Quality of life (QoL) during treatment was assessed by EORTC QLQ-C30 questionnaire.
      • There was significant improvement in QoL score for olaparib versus chemotherapy.
      • More patients receiving olaparib showed an improvement in functional subscales.
      • Only nausea/vomiting symptom score was worse with olaparib versus chemotherapy.

      Abstract

      Background

      The phase III OlympiAD study (NCT02000622) showed a statistically significant progression-free survival benefit with olaparib versus chemotherapy treatment of physician's choice (TPC) in patients with a germline BRCA mutation and human epidermal growth factor receptor 2-negative metastatic breast cancer. From this study, we report the effect of olaparib on health-related quality of life (HRQoL).

      Methods

      Patients were randomised 2:1 to olaparib monotherapy (300 mg twice daily) or single-agent TPC. The primary HRQoL end-point was mean change from baseline in the two-item global health status/QoL score determined from patient-completed European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30-item module (EORTC QLQ-C30) questionnaires and assessed using a mixed model for repeated measures. Symptoms and functioning domains, best overall response and time to deterioration of QoL were also evaluated.

      Results

      Overall questionnaire compliance rates were 93.2% for olaparib and 76.3% for TPC. Between-treatment global health status/QoL comparison showed a significant improvement in the olaparib arm versus the TPC arm, with mean change of 3.9 (standard deviation 1.2) versus −3.6 (2.2), a difference of 7.5 points (95% confidence interval [CI]: 2.48, 12.44; P = 0.0035). A higher proportion of patients in the olaparib arm showed a best overall response of ‘improvement’ in global health status/QoL (33.7% vs 13.4%). Median time to global health status/QoL deterioration was not reached in olaparib patients and was 15.3 months for TPC patients (hazard ratio: 0.44 [95% CI: 0.25, 0.77]; P = 0.004). For EORTC QLQ-C30 symptoms and functioning subscales, only nausea/vomiting symptom score was worse in the olaparib arm than in the TPC arm (across all visits compared with baseline).

      Conclusion

      HRQoL was consistently improved for patients treated with olaparib, compared with chemotherapy TPC.

      Keywords

      1. Introduction

      The management of metastatic breast cancer (mBC) is an ongoing challenge, characterised by low median overall survival of 2–3 years and 5-year survival of ∼25% [
      • Cardoso F.
      • Costa A.
      • Senkus E.
      • Aapro M.
      • André F.
      • Barrios C.H.
      • et al.
      ESO-ESMO international consensus guidelines for advanced breast cancer (ABC 3).
      ]. Patients with metastatic, triple-negative breast cancer represent an even greater treatment challenge, with median overall survival of ∼1 year [
      • Khosravi-Shahi P.
      • Cabezon-Gutierrez L.
      • Custodio-Cabello S.
      Metastatic triple negative breast cancer: optimizing treatment options, new and emerging targeted therapies.
      ]. It is now recognised that antitumour therapy for patients with mBC should be implemented with the dual goals of prolonging patient survival and optimising their quality of life (QoL) [
      • Cardoso F.
      • Costa A.
      • Senkus E.
      • Aapro M.
      • André F.
      • Barrios C.H.
      • et al.
      ESO-ESMO international consensus guidelines for advanced breast cancer (ABC 3).
      ]. In addition, routine care of patients with cancer should also include readiness of healthcare professionals to modify treatment strategies based on treatment efficacy, disease characteristics, adverse drug reactions and patient-reported outcomes (PROs; symptoms, functioning and health-related quality of life [HRQoL]) [
      • Cardoso F.
      • Costa A.
      • Senkus E.
      • Aapro M.
      • André F.
      • Barrios C.H.
      • et al.
      ESO-ESMO international consensus guidelines for advanced breast cancer (ABC 3).
      ].
      For patients with functional deficiency in BRCA1 and/or BRCA2 (BRCAm) and human epidermal growth factor receptor 2 (HER2)-negative mBC, chemotherapy may help reduce disease symptoms and prolong survival, but it is also associated with substantial toxicity that can detrimentally impact patients' QoL [
      • Avis N.E.
      • Crawford S.
      • Manuel J.
      Psychosocial problems among younger women with breast cancer.
      ,
      • Hartmann J.T.
      • Lipp H.P.
      Toxicity of platinum compounds.
      ]. Poly (ADP-ribose) polymerase (PARP) inhibitors have recently emerged as new treatment options for patients with BRCAm HER2-negative mBC [
      • Robson M.
      • Im S.A.
      • Senkus E.
      • Xu B.
      • Domchek S.M.
      • Masuda N.
      • et al.
      Olaparib for metastatic breast cancer in patients with a germline BRCA mutation.
      ,
      • Litton J.K.
      • Rugo H.S.
      • Ettl J.
      • Hurvitz S.A.
      • Gonçalves A.
      • Lee K.H.
      • et al.
      Talazoparib in patients with advanced breast cancer and a germline BRCA mutation.
      ]. The phase III OlympiAD study in patients with germline BRCAm (gBRCAm) HER2-negative mBC showed that olaparib monotherapy resulted in a statistically significant and clinically meaningful progression-free survival (PFS) benefit versus chemotherapy treatment of physician's choice (TPC): 7.0 versus 4.2 months; hazard ratio (HR): 0.58; 95% confidence interval (CI): 0.43–0.80; P < 0.001 [
      • Robson M.
      • Im S.A.
      • Senkus E.
      • Xu B.
      • Domchek S.M.
      • Masuda N.
      • et al.
      Olaparib for metastatic breast cancer in patients with a germline BRCA mutation.
      ]. Furthermore, olaparib-treated patients had a lower incidence of grade ≥3 adverse events versus TPC (36.6% vs 50.5%). Based on these results, olaparib became the first approved targeted therapy for patients with gBRCAm HER2-negative mBC who were previously treated with chemotherapy [
      • FDA
      LYNPARZA (olaparib) tablets, for oral use. Prescribing information.
      ].
      A pre-specified secondary objective of the OlympiAD study was to assess the effect of olaparib on HRQoL using the European Organisation for Research and Treatment of Cancer Quality-of-Life Questionnaire Core 30-item module (EORTC QLQ-C30) global QoL scale. We report full results regarding these longitudinal HRQoL assessments in the OlympiAD trial.

      2. Methods

      2.1 Study design and patient population

      The OlympiAD study design and patient population have been fully reported in the primary analyses [
      • Robson M.
      • Im S.A.
      • Senkus E.
      • Xu B.
      • Domchek S.M.
      • Masuda N.
      • et al.
      Olaparib for metastatic breast cancer in patients with a germline BRCA mutation.
      ]. Briefly, OlympiAD was a phase III, international, randomised, open-label study (NCT02000622) of patients aged ≥18 years with gBRCAm HER2-negative mBC (triple negative or hormone receptor positive) who had received ≤2 previous chemotherapy regimens. Patients were randomised 2:1 using an interactive voice or Web response system to olaparib tablets (300 mg twice daily) or single-agent, pre-defined, chemotherapy TPC (capecitabine, eribulin or vinorelbine – see Supplementary Materials for dosage).

      2.2 Study outcome measures

      Primary and secondary efficacy and safety outcomes are described fully in the primary manuscript [
      • Robson M.
      • Im S.A.
      • Senkus E.
      • Xu B.
      • Domchek S.M.
      • Masuda N.
      • et al.
      Olaparib for metastatic breast cancer in patients with a germline BRCA mutation.
      ]. The pre-specified PROs reported here were assessed using the paper-based 30-item EORTC QLQ-C30 questionnaire completed by the patient at baseline (before randomisation once eligibility was confirmed) and every 6 weeks until investigator-assessed objective disease progression. Patients who discontinued treatment because of toxicity continued PRO assessments until disease progression. The EORTC QLQ-C30 includes a two-item global QoL scale (global health status/QoL score); five multi-item functional scales (physical, role, emotional, cognitive and social); three multi-item symptom scales (fatigue, pain and nausea/vomiting); five single items assessing common cancer symptoms (dyspnoea, insomnia, appetite loss, constipation and diarrhoea) and a single item addressing the financial impact of disease [
      • Aaronson N.K.
      • Ahmedzai S.
      • Bergman B.
      • Bullinger M.
      • Cull A.
      • Duez N.J.
      • et al.
      The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology.
      ]. Scores for the QLQ-C30 range from 0 to 100. For global health status/QoL score and functional scales, higher scores indicate better HRQoL and level of functioning. For symptom scales, higher scores indicate greater severity of symptoms. The impact of olaparib on symptoms and HRQoL as assessed by the multi-item functional scales, multi-item symptoms scales and common cancer symptoms, were exploratory analyses.
      The primary HRQoL end-point compared mean change from baseline between treatment arms in the two-item global health status/QoL score across all visits. Secondarily, we assessed the proportion of patients in each arm who experienced a clinically meaningful increase (improvement) or decrease (deterioration) in global health status/QoL score (defined as ≥10-point change from baseline) [
      • Osoba D.
      • Rodrigues G.
      • Myles J.
      • Zee B.
      • Pater J.
      Interpreting the significance of changes in health-related quality-of-life scores.
      ]. Best overall HRQoL response was defined as the best HRQoL response the patient achieved from randomisation to disease progression. Best HRQoL response was categorised as ‘improved’, ‘no change’, ‘deterioration’ or ‘other’ according to the following criteria: ‘improved’ – two visit responses of ‘improved’ sustained for ≥21 days with no intervening response of ‘deterioration’; ‘no change’ – two visit responses of either ‘no change’ or ‘improved’ and ‘no change’ ≥21 days apart with no intervening response of ‘deterioration’; ‘deterioration’ – a visit response of ‘deterioration’ without a response of ‘improved’ or ‘no change’ within 21 days; ‘other’ – patients who met the criteria for ‘improved’, ‘no change’ or ‘deterioration’. For additional information on continued treatment effectiveness, an analysis of the number of patients remaining on treatment for ≥6 months and time to subsequent therapies are shown in Supplementary Materials.

      2.3 Statistical analysis

      Sample size determination has been reported in the primary manuscript [
      • Robson M.
      • Im S.A.
      • Senkus E.
      • Xu B.
      • Domchek S.M.
      • Masuda N.
      • et al.
      Olaparib for metastatic breast cancer in patients with a germline BRCA mutation.
      ]. Questionnaire compliance and completion data were analysed overall and by study visit and summarised by treatment. All PRO data were analysed using the full analysis set on an intention-to-treat basis that included all randomised patients with a baseline and ≥1 post-baseline assessment. For the primary HRQoL analysis, a linear mixed-model repeated-measures (MMRM) analysis adjusting for score at baseline, time and treatment-by-time interaction was used to estimate the cumulative effect of olaparib versus TPC on global health status/QoL. MMRM methodology uses observed data to implicitly impute unobserved data; by including covariates in the model, the analysis assumes that patients with missing data would behave similarly to other patients with similar values for these covariates (e.g. treatment group) had they not missed the assessment. The analysis included all post-baseline visits up to the last scheduled visit in which ≥20 patients in each treatment arm had an evaluable score. Differences between treatment groups were compared using adjusted mean estimates per treatment group and corresponding 95% CIs and P values.
      Best overall global health status/QoL response rates (improvement, no change or deterioration) were summarised descriptively for patients. Multi-item functional and symptom subscales and single-item cancer symptoms were exploratory variables. For global health status/QoL and multi-item functional subscales, best overall response was evaluated in patients with baseline score ≥10. For multi-item symptom subscales and single-item symptom scales, best overall response was evaluated in patients with baseline score ≤90.
      Time to deterioration (TTD) of global health status/QoL (planned analysis) and functional and symptom scales (post hoc analysis) was defined as time from randomisation until the date of a clinically important deterioration in the global health status/QoL score or functional and symptom scores. For multi-item functional and global health status/QoL subscales, deterioration included patients with baseline score ≥10; for symptom subscales/single items, deterioration included patients with baseline score ≤90. Clinically important deterioration was defined as a decrease from baseline of ≥10 points (or an increase from baseline of ≥10 points for the symptom scales) that was sustained at the next scheduled visit. Patients who did not experience deterioration by the time of progression (PRO data beyond progression were not collected) were censored at the time of the last-available PRO assessment. Data were analysed using a log-rank test, with HR and 95% CI generated from the log-rank test statistics. The Kaplan–Meier method was used to estimate medians.

      3. Results

      3.1 Study population characteristics

      In total, 302 patients (intention-to-treat population) were randomly assigned to treatment (olaparib, n = 205; TPC, n = 97) between 7 April 2014 and 27 November 2015. Of the 97 TPC patients, 91 received study treatment (capecitabine, n = 41; eribulin, n = 34; vinorelbine, n = 16). A participant flow diagram (Fig. 1) and baseline demographic characteristics, which were well balanced between the two treatment groups, have been described [
      • Robson M.
      • Im S.A.
      • Senkus E.
      • Xu B.
      • Domchek S.M.
      • Masuda N.
      • et al.
      Olaparib for metastatic breast cancer in patients with a germline BRCA mutation.
      ]. Mean (standard deviation [SD]) global health status/QoL score (scale range: 0–100; higher scores indicate higher levels of function and QoL) at baseline was 63.2 (21.0) for the olaparib treatment arm and 63.3 (21.2) for TPC (capecitabine 62.3 [19.6], eribulin 59.1 [22.1], vinorelbine 67.7 [21.5]) [
      • Robson M.
      • Im S.A.
      • Senkus E.
      • Xu B.
      • Domchek S.M.
      • Masuda N.
      • et al.
      Olaparib for metastatic breast cancer in patients with a germline BRCA mutation.
      ]. Multi-item symptom scores at baseline in both treatment arms were similar to those cited for recurrent or mBC in the EORTC QLQ-C30 reference value manual [
      • Scott N.W.
      • Fayers P.M.
      • Aaronson N.K.
      • Bottomley A.
      • de Graeff A.
      • Groenvold M.
      • et al.
      ].
      Fig. 1
      Fig. 1Participant flow diagram. bid, twice daily; TPC, treatment of physician's choice.

      3.2 Questionnaire compliance/completion

      Compliance and completion rates for EORTC QLQ-C30 at baseline were >95% in both arms (Table 1). Completion rates declined faster in the TPC arm.
      Table 1Compliance and completion rates for the EORTC QLQ-C30 questionnaire.
      Time pointExpected forms, nCompliance rate,
      Compliance rate for each treatment arm was defined as follows: (the number of evaluable questionnaires [defined as a questionnaire with enough responses to score at least one scale/domain])/(the number of patients expected to complete questionnaires at each visit [i.e. patients still under PRO follow-up]) × 100%.
      %
      Completion rate,
      Completion rate for each treatment arm was defined as follows: [(the number of received questionnaires)/(PRO analysis population)] × 100%.
      %
      OlaparibTPCOlaparib (n = 205)TPC (n = 97)Olaparib (n = 205)TPC (n = 97)
      Baseline2059799.095.999.095.9
      Visit 6 (week 6)2059687.371.987.371.1
      Visit 8 (week 12)1847387.071.278.053.6
      Visit 10 (week 18)1675084.470.068.836.1
      Visit 12 (week 24)1363684.663.956.123.7
      Visit 14 (week 30)1153084.463.347.319.6
      All visits (overall)
      For overall, patients were counted as received/evaluable if they had a received/evaluable baseline and at least one received/evaluable post-baseline form.
      2059793.276.393.276.3
      EORTC QLQ-C30, European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30-item module; PRO, patient-reported outcome; TPC, treatment of physician's choice.
      Note that 19 assessments are included whereby the questionnaire was administered to a caregiver, relative or unspecified person, rather than directly to the patient.
      a Compliance rate for each treatment arm was defined as follows: (the number of evaluable questionnaires [defined as a questionnaire with enough responses to score at least one scale/domain])/(the number of patients expected to complete questionnaires at each visit [i.e. patients still under PRO follow-up]) × 100%.
      b Completion rate for each treatment arm was defined as follows: [(the number of received questionnaires)/(PRO analysis population)] × 100%.
      c For overall, patients were counted as received/evaluable if they had a received/evaluable baseline and at least one received/evaluable post-baseline form.

      3.3 Global health status/QoL

      The between-treatment comparison (average over time) based on an MMRM model showed a clinically significant improvement from baseline in mean (SD) global health status/QoL score for olaparib versus TPC (3.9 [1.2] vs −3.6 [2.2]; difference 7.5 [95% CI: 2.48–12.44]; P = 0.0035) [
      • Robson M.
      • Im S.A.
      • Senkus E.
      • Xu B.
      • Domchek S.M.
      • Masuda N.
      • et al.
      Olaparib for metastatic breast cancer in patients with a germline BRCA mutation.
      ]. The adjusted mean change from baseline over time (Fig. 2) showed that, for each visit, patients in the olaparib arm had an improvement in mean global health status/QoL score, whereas patients in the TPC arm had a decline.
      Fig. 2
      Fig. 2Adjusted mean (SD) change from baseline in EORTC QLQ-C30 global health status/quality of life score across time points in patients in the olaparib and TPC arms. bid, twice daily; EORTC QLQ-C30, European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30-item module; SD, standard deviation; TPC, treatment of physician's choice. A higher score represents better overall health-related quality of life. Note that data are restricted to visits with at least 20 patients in each treatment arm.
      Median TTD in global health status/QoL (≥10 points) was not reached in the olaparib arm and was 15.3 months in the TPC arm (HR: 0.44; 95% CI: 0.25–0.77; P = 0.004; Fig. 3) [
      • Robson M.
      • Im S.A.
      • Senkus E.
      • Xu B.
      • Domchek S.M.
      • Masuda N.
      • et al.
      Olaparib for metastatic breast cancer in patients with a germline BRCA mutation.
      ]. Global scores had deteriorated in fewer patients in the olaparib arm versus TPC at 6 months (18.5% vs 38.8%) and 12 months (36.0% vs 46.5%). Censoring of most olaparib-treated patients because they did not meet the 10-point threshold criterion for deterioration means that these TTD data should be interpreted with caution (see Discussion).
      Fig. 3
      Fig. 3Kaplan–Meier plot of time to deterioration in EORTC QLQ-C30 global health status/quality of life at progression. bid, twice daily; CI, confidence interval; EORTC QLQ-C30, European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30-item module; HR, hazard ratio; NC, not calculable; QoL, quality of life; TPC, treatment of physician's choice. Deterioration was defined as the onset of a ≥10-point decrease in global health status/QoL score from baseline. Patients whose global health status/QoL score did not show a clinically important deterioration and who were alive at the time of analysis, or who experienced deterioration in global health status/QoL or death after two or more missed assessments, were censored at the latest evaluable EORTC QLQ-C30 assessment.
      More patients in the olaparib treatment arm showed an improvement in best overall response rates for global health status/QoL versus TPC (33.7% vs 13.4%), and more patients in the TPC arm showed a deterioration in global health status/QoL versus olaparib (20.6% vs 11.7%) (Fig. 4).
      Fig. 4
      Fig. 4Best overall response (improvement, no change and deterioration) rates for EORTC QLQ-C30 global health status/quality of life. QoL, quality of life; TPC, treatment of physician's choice. Best overall response of improvement and deterioration were defined as a ≥10-point change from baseline and sustained for at least 21 days (for ‘improvement’, without an intervening visit response of ‘deterioration’; for ‘deterioration’, without an intervening response of ‘improved’ or ‘no change’). *‘Other’ includes patients with a best response of ‘other’, defined as patients who did not demonstrate improvement, no change or deterioration, and those with missing follow-up data or non-evaluable patients (owing to baseline scores being outside the limit). Patients typically classed as ‘other’, for example, did not have a sustained response observed at the next visit but had one visit response of improvement, followed by a deterioration, then a visit with no change.

      3.4 Functional subscales

      Kaplan–Meier estimates of TTD (before or at progression) showed that olaparib delayed TTD versus TPC for each of the five functional subscales (Fig. 5).
      Fig. 5
      Fig. 5Kaplan–Meier estimates of time to deterioration for EORTC QLQ-C30 functional subscales before or at progression. (a) Physical; (b) role; (c) social; (d) cognitive and (e) emotional. bid, twice daily; CI, confidence interval; EORTC QLQ-C30, European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30-item module; HR, hazard ratio; NC, not calculable; TPC, treatment of physician's choice.
      Fig. 5
      Fig. 5Kaplan–Meier estimates of time to deterioration for EORTC QLQ-C30 functional subscales before or at progression. (a) Physical; (b) role; (c) social; (d) cognitive and (e) emotional. bid, twice daily; CI, confidence interval; EORTC QLQ-C30, European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30-item module; HR, hazard ratio; NC, not calculable; TPC, treatment of physician's choice.
      The best overall response data for the functional subscales support the results for global health status/QoL and highlight that, across all five functional subscales, more patients experienced an improvement in functional subscales in the olaparib treatment arm versus TPC (Supplementary Fig. 1A); concordantly, more patients experienced deterioration in functional subscales in the TPC arm versus olaparib.

      3.5 Symptom subscales

      For fatigue, pain and nausea/vomiting multi-item symptom scales (decrease in score indicates improvement in symptoms), as well as single-item symptom scales, a higher proportion of patients in the olaparib treatment arm experienced improvements versus TPC (Supplementary Fig. 1B and 1C).
      Adjusted mean changes from baseline in symptom scores across all visits (Fig. 6) showed improvements in fatigue, pain, dyspnoea, insomnia, loss of appetite, constipation and diarrhoea scores among patients receiving olaparib versus patients receiving TPC. Only mean nausea/vomiting scores were better (relative to baseline) in the TPC arm versus olaparib. Apart from nausea/vomiting, the HR for TTD favoured olaparib versus TPC across all symptom scales (Supplementary Fig. 2).
      Fig. 6
      Fig. 6Estimated difference (olaparib vs TPC) in adjusted mean change from baseline in EORTC QLQ-C30 symptom subscale scores (all visits). EORTC QLQ-C30, European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30-item module; TPC, treatment of physician's choice.

      3.6 Time to subsequent therapies

      Results of the analyses of time to subsequent therapies are presented in Supplementary Materials.

      4. Discussion

      Optimising HRQoL is a key component in the guidelines for treatment and management of mBC as treatment is usually not curative but palliative [
      • Cardoso F.
      • Costa A.
      • Senkus E.
      • Aapro M.
      • André F.
      • Barrios C.H.
      • et al.
      ESO-ESMO international consensus guidelines for advanced breast cancer (ABC 3).
      ]. Our findings show that the efficacy benefits reported with olaparib among patients with gBRCAm HER2-negative mBC are accompanied by improvements in symptoms, functioning and HRQoL assessed using the EORTC QLQ-C30.
      Adjusted mean change from baseline in global health status/QoL score across all visits revealed greater improvements with olaparib versus TPC. Overall, HRQoL in patients receiving olaparib improved during time on treatment versus patients receiving TPC, in whom deterioration in HRQoL over time was observed. The difference between arms is consistent with a small, clinically significant improvement (7.5-point difference lies within the 5- to 10-point change designated as small [moderate, 10–20; large, >20]) on a group-level assessment [
      • Osoba D.
      • Rodrigues G.
      • Myles J.
      • Zee B.
      • Pater J.
      Interpreting the significance of changes in health-related quality-of-life scores.
      ,
      • Cocks K.
      • King M.T.
      • Velikova G.
      • de Castro G.
      Evidence-based guidelines for interpreting change scores for the European organisation for the research and treatment of cancer quality of life questionnaire Core 30.
      ]. TTD was also extended in the olaparib arm versus TPC. However, the median TTD findings should be interpreted cautiously as PRO data were collected only until disease progression, and most patients had not reached the 10-point threshold criterion for deterioration at this time. Patients who did not meet the deterioration criteria were censored at the time of their last global health status/QoL assessment. Given these limitations, the best overall response represents a more instructive assessment to determine the impact of treatments on HRQoL. The findings of the best overall response showed that olaparib-treated patients were almost three times as likely as those receiving TPC to report a clinically significant improvement in global health status/QoL.
      EORTC QLQ-C30 functional and symptom subscale scores were generally improved in the olaparib arm versus TPC. Marked improvements in functioning favoured olaparib over TPC, as measured on the physical, cognitive and emotional subscales. For each symptom subscale, more patients in the olaparib group showed a best overall response of a clinically meaningful improvement in symptom score versus TPC. Single symptoms included in the scale are considered as being important to patients with breast cancer. Our findings showed particular benefits of olaparib over TPC in terms of improved pain and dyspnoea. For patients who experienced deterioration in symptoms, this was typically greater for those receiving TPC versus olaparib.
      Differences in adjusted mean change from baseline in symptom scores for nausea/vomiting favoured TPC. This is a known effect of olaparib that has been shown to be greater during treatment initiation [
      • Banerjee S.
      • Ledermann J.
      • Matulonis U.
      • Molife L.R.
      • Friedlander M.
      • Fielding A.
      • et al.
      Management of nausea and vomiting during treatment with the capsule (CAP) and tablet (TAB) formulations of the PARP inhibitor olaparib.
      ]. Indeed, a high proportion of patients receiving olaparib reported an improvement in the nausea/vomiting subscale over time. Previously reported safety data from the OlympiAD trial showed that more patients experienced nausea/vomiting with olaparib versus TPC, but no grade ≥3 events were reported and discontinuation rates were low [
      • Robson M.
      • Im S.A.
      • Senkus E.
      • Xu B.
      • Domchek S.M.
      • Masuda N.
      • et al.
      Olaparib for metastatic breast cancer in patients with a germline BRCA mutation.
      ,
      • Robson M.E.
      • Tung N.
      • Conte P.
      • Im S.A.
      • Senkus E.
      • Xu B.
      • et al.
      OlympiAD final overall survival and tolerability results: olaparib versus chemotherapy treatment of physician's choice in patients with a germline BRCA mutation and HER2-negative metastatic breast cancer.
      ]. We found that although a higher proportion of olaparib-treated patients reported a best overall response of improvement in nausea/vomiting, differences in adjusted mean change from baseline in symptom scores for nausea/vomiting favoured TPC. These data are hard to reconcile but suggest that patients whose symptoms do not improve experience a greater decrease in score. In clinical practice, supportive measures such as early antiemetic administration may be appropriate to manage events of nausea and vomiting, thus minimising olaparib dose adjustments/discontinuations. The improvements in HRQoL in the olaparib arm versus TPC are consistent with the lower rate of discontinuations because of adverse events (4.9% vs 7.7%) and the longer median treatment duration (8.2 [range 0.5–28.7] vs 3.4 [0.7–23.0] months) reported previously in the OlympiAD trial [
      • Robson M.
      • Im S.A.
      • Senkus E.
      • Xu B.
      • Domchek S.M.
      • Masuda N.
      • et al.
      Olaparib for metastatic breast cancer in patients with a germline BRCA mutation.
      ]. The supportive analyses of time to subsequent therapies (see Supplementary Materials) further expand on these data, showing that more patients in the olaparib arm remained on treatment for ≥6 months versus TPC (60.0% vs 27.5%), and median time to first and second subsequent therapy was longer with olaparib versus TPC (9.4 vs 4.2 months and 14.3 vs 10.5 months, respectively); time on assigned treatment has been recognised as a meaningful clinical trial end-point in settings whereby a disease course involves multiple rounds of subsequent treatment after first progression [
      • Matulonis U.A.
      • Oza A.M.
      • Ho T.W.
      • Ledermann J.A.
      Intermediate clinical endpoints: a bridge between progression-free survival and overall survival in ovarian cancer trials.
      ].
      Limitations of the study include the open-label trial design, which has the potential to introduce patient biases. In addition, the absence of post-progression PRO data collection may have underestimated the impact of olaparib on HRQoL given that patients receiving TPC had disease progression earlier than those receiving olaparib, and progression would be expected to result in a further decrease in HRQoL. However, this would also be dependent on the HRQoL impact of post-progression treatment. Questionnaire compliance and completion rates were lower in the TPC arm versus olaparib (76.3% vs 93.2%), and completion rates also declined faster in the TPC arm because of earlier progression and cessation of EORTC QLQ-C30 data collection. This higher rate of missing data may also have biased the results of the analysis. The direction of any bias on the effect of treatment is unclear as it depends on whether missing data would have improved or worsened PRO measures for the TPC arm. However, the selective aspect of completing EORTC QLQ-C30 data collection is likely to be largely explained by observed covariates included in the MMRM analysis, i.e. treatment arm and baseline HRQoL score, which assumes that patients with missing data would behave similarly to other patients in the same treatment group, and with similar covariate values, had they not missed the assessment. Studies have shown that when such observed covariates are included, the adjusted estimates are robust even when data are not missing completely at random [
      • Sloan J.A.
      • Dueck A.C.
      • Erickson P.A.
      • Guess H.
      • Revicki D.A.
      • Santanello N.C.
      Analysis and interpretation of results based on patient-reported outcomes.
      ].
      Favourable PROs have also been reported with another PARP inhibitor among patients with gBRCAm advanced breast cancer [
      • Litton J.K.
      • Rugo H.S.
      • Ettl J.
      • Hurvitz S.A.
      • Gonçalves A.
      • Lee K.H.
      • et al.
      Talazoparib in patients with advanced breast cancer and a germline BRCA mutation.
      ]. Improvement in the global health score from baseline with talazoparib in the EMBRACA trial (3.0) was similar to that reported here (3.9). Our findings and those of EMBRACA provide further support of the beneficial effects of PARP inhibitors on HRQoL versus chemotherapy in patients with mBC. Based on these data, recently updated advanced breast cancer (ABC4) and National Comprehensive Cancer Network guidelines recommend PARP inhibitors as a treatment option for BRCA-associated HER2-negative mBC treated with an anthracycline with or without a taxane in the adjuvant and/or metastatic setting [
      • Cardoso F.
      • Senkus E.
      • Costa A.
      • Papadopoulos E.
      4th ESO-ESMO international consensus guidelines for advanced breast cancer (ABC 4).
      ,
      • National Comprehensive Cancer Network
      NCCN guidelines for the treatment of cancer by site.
      ].

      5. Conclusions

      The planned and post hoc analyses of data we report from the phase III OlympiAD study suggest that olaparib treatment can lead to improvements in the symptoms, functioning and HRQoL of patients with gBRCAm HER2-negative mBC versus chemotherapy TPC. PRO results will be valuable to both physicians and patients when considering the clinical benefits of PARP inhibitor treatment in mBC, versus chemotherapy.

      Data access, responsibility and analysis

      Data underlying the findings described in this manuscript may be obtained in accordance with AstraZeneca's data sharing policy described at https://astrazenecagrouptrials.pharmacm.com/ST/Submission/Disclosure.
      M.R., W.B. and C.G. had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. W.B. is the statistical author who has responsibility for the statistical analysis.

      Conflict of interest statement

      M.R. reports receiving honoraria (advisory) from AstraZeneca and has served a consulting or advisory role for AstraZeneca, Daiichi Sankyo (uncompensated), McKesson, Merck (uncompensated) and Pfizer (uncompensated). Research funding has been received from AbbVie (institution), AstraZeneca (institution), Invitae (institution, in-kind), Medivation (institution), Myriad Genetics (institution, in-kind), Pfizer (institution) and Tesaro (institution). Travel or accommodation expenses have been received from AstraZeneca, and other transfer of value items are from AstraZeneca (editorial services) and Pfizer (editorial services). K.J.R. reports intellectual property interests for herself and her spouse regarding a discovery or technology relating to health or medicine. S-.A.I. has served a consultant role for AstraZeneca , Novartis , Spectrum , Hanmi, Pfizer  and Roche and reports receiving research funding from AstraZeneca . E.S. reports receiving honoraria from Amgen, Astellas, AstraZeneca, Bayer, Bristol-Myers Squibb, Celgene, Clinigen, Egis, Eli Lilly, Janssen, Novartis, Pfizer, Pierre Fabre, prIME, Roche and Teva; travel support from Amgen, AstraZeneca, Egis, Novartis, Pfizer and Roche and research funding from Amgen, Astellas, AstraZeneca, Bayer, Bristol-Myers Squibb, Boehringer, Eli Lilly, Janssen, Merck, Novartis, Pfizer, Roche and Samsung. B.X. reports no conflicts of interest. S.M.D. reports receiving honoraria from AstraZeneca , Clovis Oncology  and Bristol-Myers Squibb , and research funding to the University of Pennsylvania from AstraZeneca and Clovis Oncology . N.M. reports receiving honoraria from Chugai Pharma, AstraZeneca , Pfizer  and Takeda, and institutional research funding from Chugai Pharma, AstraZeneca , Kyowa Hakka Kirin, MSD , Novartis , Pfizer , Eli- Lilly  and Daiichi Sankyo. W.L. reports no conflicts of interest. N.T. reports consulting for AstraZeneca and receiving grant support from Myriad Genetics and Ambry Genetics. A.A . reports receiving fees for serving on an advisory board from Roche and Syndax Pharmaceuticals and her spouse holding stock options in AstraZeneca . S.D. reports receiving honoraria from Novartis , Roche , AstraZeneca , Pfizer , GE Healthcare and Puma Biotechnology; consulting fees for Novartis , Roche , Pfizer , AstraZeneca and Puma Biotechnology; research funding from Novartis , Roche , AstraZeneca , Pfizer and Puma Biotechnology; and travel and accommodation expenses from Pfizer , AstraZeneca and Novartis . W.B., C.G., A.D. and R.H. are employees of AstraZeneca. P.C . reports being a member of speakers’ bureaus of Roche , Novartis , AstraZeneca and GlaxoSmithKline and receiving travel and accommodation expenses from Novartis , GlaxoSmithKline and Celgene, and research funding from Roche , Novartis and Merck Serono.

      Acknowledgements

      Medical writing assistance was provided by Debbi Gorman, PhD, and Martin Goulding, DPhil, from Mudskipper Business Ltd, funded by AstraZeneca and MSD .

      Appendix A. Supplementary data

      The following is the supplementary data to this article:

      Funding

      This study was supported by AstraZeneca and Merck Sharp & Dohme Corp, a subsidiary of Merck & Co, Inc, Kenilworth, NJ, USA (MSD), who are co-developing olaparib. The study sponsor, AstraZeneca , was involved in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation, review or approval of the manuscript and decision to submit the manuscript for publication.

      References

        • Cardoso F.
        • Costa A.
        • Senkus E.
        • Aapro M.
        • André F.
        • Barrios C.H.
        • et al.
        ESO-ESMO international consensus guidelines for advanced breast cancer (ABC 3).
        Breast. 2017; 31: 244-259
        • Khosravi-Shahi P.
        • Cabezon-Gutierrez L.
        • Custodio-Cabello S.
        Metastatic triple negative breast cancer: optimizing treatment options, new and emerging targeted therapies.
        Asia Pac J Clin Oncol. 2018; 14: 32-39
        • Avis N.E.
        • Crawford S.
        • Manuel J.
        Psychosocial problems among younger women with breast cancer.
        Psycho Oncol. 2004; 13: 295-308
        • Hartmann J.T.
        • Lipp H.P.
        Toxicity of platinum compounds.
        Expert Opin Pharmacother. 2003; 4: 889-901
        • Robson M.
        • Im S.A.
        • Senkus E.
        • Xu B.
        • Domchek S.M.
        • Masuda N.
        • et al.
        Olaparib for metastatic breast cancer in patients with a germline BRCA mutation.
        N Engl J Med. 2017; 377: 523-533
        • Litton J.K.
        • Rugo H.S.
        • Ettl J.
        • Hurvitz S.A.
        • Gonçalves A.
        • Lee K.H.
        • et al.
        Talazoparib in patients with advanced breast cancer and a germline BRCA mutation.
        N Engl J Med. 2018; 379: 753-763
        • FDA
        LYNPARZA (olaparib) tablets, for oral use. Prescribing information.
        (Available at:)
        • Aaronson N.K.
        • Ahmedzai S.
        • Bergman B.
        • Bullinger M.
        • Cull A.
        • Duez N.J.
        • et al.
        The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology.
        J Natl Cancer Inst. 1993; 85: 365-376
        • Osoba D.
        • Rodrigues G.
        • Myles J.
        • Zee B.
        • Pater J.
        Interpreting the significance of changes in health-related quality-of-life scores.
        J Clin Oncol. 1998; 16: 139-144
        • Scott N.W.
        • Fayers P.M.
        • Aaronson N.K.
        • Bottomley A.
        • de Graeff A.
        • Groenvold M.
        • et al.
        EORTC QLQ-C30 reference values.
        (Available at:)
        • Cocks K.
        • King M.T.
        • Velikova G.
        • de Castro G.
        Evidence-based guidelines for interpreting change scores for the European organisation for the research and treatment of cancer quality of life questionnaire Core 30.
        Eur J Cancer. 2012; 48: 1713-1721
        • Banerjee S.
        • Ledermann J.
        • Matulonis U.
        • Molife L.R.
        • Friedlander M.
        • Fielding A.
        • et al.
        Management of nausea and vomiting during treatment with the capsule (CAP) and tablet (TAB) formulations of the PARP inhibitor olaparib.
        in: European cancer congress. Vienna, Austria, 25–29 September. 2015 (abst 2759)
        • Robson M.E.
        • Tung N.
        • Conte P.
        • Im S.A.
        • Senkus E.
        • Xu B.
        • et al.
        OlympiAD final overall survival and tolerability results: olaparib versus chemotherapy treatment of physician's choice in patients with a germline BRCA mutation and HER2-negative metastatic breast cancer.
        Ann Oncol. 2019; 30: 558-566
        • Matulonis U.A.
        • Oza A.M.
        • Ho T.W.
        • Ledermann J.A.
        Intermediate clinical endpoints: a bridge between progression-free survival and overall survival in ovarian cancer trials.
        Cancer. 2015; 121: 1737-1746
        • Sloan J.A.
        • Dueck A.C.
        • Erickson P.A.
        • Guess H.
        • Revicki D.A.
        • Santanello N.C.
        Analysis and interpretation of results based on patient-reported outcomes.
        Value Health. 2007; 10: S106-S115
        • Cardoso F.
        • Senkus E.
        • Costa A.
        • Papadopoulos E.
        4th ESO-ESMO international consensus guidelines for advanced breast cancer (ABC 4).
        Ann Oncol. 2018; 29: 1634-1657
        • National Comprehensive Cancer Network
        NCCN guidelines for the treatment of cancer by site.
        (Available at:)