PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Lancet Oncol. Author manuscript; available in PMC 2016 July 1.
Published in final edited form as:
PMCID: PMC4562429
NIHMSID: NIHMS717416

Adjuvant exemestane versus tamoxifen in premenopausal women with early breast cancer undergoing ovarian suppression: patient-reported outcomes in the TEXT and SOFT randomised trials

Abstract

Background

Efficacy analysis of the combined IBCSG TEXT and SOFT trials showed a significant disease-free survival benefit for exemestane plus ovarian function suppression (OFS) compared with tamoxifen +OFS. We present patient-reported outcomes from these trials.

Methods

Between 7 November 2003 and 7 April 2011, 4717 premenopausal patients with hormone-receptor positive breast cancer were enrolled in TEXT or SOFT to receive unblinded adjuvant treatment with 5 years of exemestane+OFS or tamoxifen+OFS. Chemotherapy use was optional. Randomization was performed via IBCSG’s internet-based system with the use of permuted blocks and was stratified by chemotherapy use and lymph nodes status. Patients completed a quality of life (QoL) form including several global and symptom-specific indicators at baseline, every 6 months for 24 months, then annually during years 3 to 6. Differences in change of QoL from baseline between the two treatments were tested at short-, mid-, and long-term using mixed-models for repeated measures, for each trial with and without chemotherapy and overall. The analysis was intention-to-treat using treatment as randomly assigned. At the time of analysis, the median follow-up was 5·7 years (IQR, 3·7 to 6·9 years), with treatment and follow-up of patients ongoing.

Findings

Patients reported considerable worsening from baseline in key endocrine symptoms. Those on tamoxifen+OFS were more affected by hot flushes and sweats over five years than those on exemestane+OFS, although these symptoms improved. Patients on exemestane+OFS reported more vaginal dryness, greater loss of sexual interest, and difficulties becoming aroused. These differences persisted over time. An increase in bone/joint pain was more pronounced, particularly in the short-term, in patients on exemestane+OFS. Changes of global QoL indicators from baseline were small and similar between treatments over the whole treatment period.

Interpretation

Overall, from a QoL perspective, there is no strong indication to favor either exemestane+OFS or tamoxifen+OFS. The differential effects of the two treatments on endocrine symptoms burden need to be addressed with patients individually.

Funding

TEXT and SOFT receive financial support for trial conduct from Pfizer, the International Breast Cancer Study Group and the US National Cancer Institute. Pfizer and Ipsen provide drug supply. See Acknowledgment for grants and grant numbers.

Keywords: adjuvant therapy, aromatase inhibitor, breast cancer, endocrine therapy, exemestane, GnRH agonist, ovarian function suppression, premenopausal, quality of life, sexual functioning, tamoxifen, triptorelin

INTRODUCTION

In premenopausal patients with hormone receptor-positive breast cancer, the TEXT and SOFT combined analysis showed a significant disease-free survival benefit for exemestane plus ovarian function suppression (OFS) compared with tamoxifen+OFS, providing a new treatment option for premenopausal women who receive OFS as part of adjuvant endocrine therapy.1 Whether or not to give chemotherapy was decided by the physician and patient. In TEXT, chemotherapy was administered concurrently with OFS after randomization. In SOFT, patients who received chemotherapy were eligible for randomization only if they remained premenopausal after completion.2

Comparisons of tamoxifen with an aromatase inhibitor have been extensively studied in postmenopausal patients. The arimidex, tamoxifen, alone or in combination (ATAC) trial,3,4 and the intergroup exemestane study (IES)5 assessed patient-reported symptoms related to endocrine therapy. These trials revealed no major effect on overall quality of life (QoL) and total symptom scores according to endocrine agents. Patients reported hot flushes as the most prevalent side-effect but indicated no differences between tamoxifen and anastrozole3,4 or exemestane.5 Self-reported gynecological symptoms (i.e. vaginal dryness, diminished libido, pain with intercourse) were more frequent with anastrozole compared to tamoxifen, whereas dizziness, cold sweats and vaginal discharge were more frequent with tamoxifen.3,4 Comparisons between exemestane and tamoxifen revealed no differences for any of the endocrine symptoms except for vaginal discharge, which was greater with tamoxifen up to 24 months following 2–3 years of treatment with tamoxifen.5 Patient-reported bone or joint pain was not assessed in these trials.

Placebo-controlled breast cancer prevention trials in postmenopausal women showed more vasomotor6,7 and gynecologic symptoms as well as sexual problems for tamoxifen6 and worse menopausal-related QoL for exemestane8. In the recent mammary prevention 3 (MAP.3) trial, exemestane had small negative effects on vasomotor symptoms, sexual symptoms, and pain, which occurred mainly in the first 6 months to 2 years after random assignment. No treatment differences were observed in general QoL domains.9

Little is known about the influence of chemotherapy on patients’ perception of endocrine symptoms. In the Zoladex® in premenopausal patients (ZIPP) trial, the endocrine therapy (goserelin, goserelin plus tamoxifen or tamoxifen alone) had differential effects on patient-reported symptoms only in those patients who did not receive chemotherapy.10

The QoL analysis in the TEXT and SOFT trials was conducted in parallel to the efficacy analysis1 to elucidate differential effects of exemestane as compared to tamoxifen on QoL among patients who are treated with OFS during adjuvant endocrine therapy.

METHODS

Study design and participants

Details of the trials have been described elsewhere.1,2 Eligibility for both trials included operable hormone receptor-positive breast cancer and premenopausal status, defined by regular menses without exogenous hormones during the prior six months and/or estradiol level in the premenopausal range. All TEXT patients and SOFT patients who did not receive chemotherapy were randomized within twelve weeks of definitive surgery; SOFT patients who received (neo)adjuvant chemotherapy were randomized within eight months of completing chemotherapy, as soon as a premenopausal estradiol level was confirmed.

TEXT was designed to evaluate five years of exemestane with the gonadotropin-releasing hormone agonist triptorelin versus tamoxifen with triptorelin for women who received OFS from the start of adjuvant therapy. Eligible women were randomized 1:1 to five years of exemestane 25mg orally daily plus triptorelin 3·75mg by intramuscular injection every 28 days, or to five years of tamoxifen 20mg orally daily plus triptorelin. As an alternative method of OFS, bilateral oophorectomy or ovarian irradiation could be performed after at least 6 months of triptorelin. Chemotherapy was optional, and if administered, was started concomitantly with triptorelin; oral endocrine therapy started after completion of chemotherapy. If chemotherapy was not administered, then oral endocrine therapy started 6 to 8 weeks after initiation of triptorelin, to allow for a prior decline in ovarian estrogen production.

SOFT was designed to evaluate five years of exemestane+OFS versus tamoxifen+OFS versus tamoxifen for women who remained premenopausal within eight months after completion of pre- or post-operative adjuvant chemotherapy, or for whom adjuvant tamoxifen alone was selected as suitable treatment. Eligible women were randomized 1:1:1 to exemestane+OFS (triptorelin, bilateral oophorectomy or ovarian irradiation), tamoxifen+OFS, or tamoxifen. Results of patients randomized to tamoxifen alone are not included in this report. Consistent with this design, patients in SOFT (but not TEXT) were allowed to receive adjuvant oral endocrine therapy before randomization.

In TEXT and SOFT, protocol-assigned endocrine therapy was to cease 5 years from randomization with annual follow-up thereafter unless the patient withdrew consent for further participation. At this point in follow-up, 30% of patients continued on some or all protocol-assigned treatments, 56% of patients had completed treatment and 14% had stopped all protocol-assigned treatments early (16% exemestane+OFS and 11% tamoxifen+OFS).1 The ethics committees of each participating center approved the study protocols and all patients gave written informed consent.

Randomisation and masking

In both trials, random allocation of unblinded treatment assignment was generated and performed via IBCSG’s internet-based system with the use of permuted blocks and was stratified by chemotherapy use (yes/no) and lymph nodes status (N0/N1+). The Participating Center Principal Investigator or designee accessed the randomisation system.

Procedures: QoL assessment

Patients completed the IBCSG QoL Core Form11 and a trial-specific module at baseline, every 6 months for 2 years, and then annually in years 3 through 6. The forms were to be completed at the clinic, before diagnostic procedures (exception: baseline) or treatment was given and regardless of disease status. The forms were faxed to the IBCSG Data Management Center in Amherst, NY. The baseline assessment in TEXT was prior to any adjuvant therapy, whereas some SOFT patients had received chemotherapy and initiated oral endocrine therapy. Eligibility exceptions were cognitive or physical impairment that interfered with the QoL assessment, or inability to read any of the 26 languages available on the core form. The core form includes global indicators for physical well-being,12 mood,13 coping effort,14 and subjective health estimation,15 and five indicators specific to symptoms and side-effects.16 The module consisted of 12 endocrine symptom indicators, one global indicator for treatment burden17 and a question regarding sexual activity during the past 6 months (yes/no). The selection of symptoms was based on the findings of two breast cancer prevention trials6,7 including a symptom checklist developed for the National Surgical Adjuvant Breast and Bowel Project Breast Cancer Prevention Trial18,19 subsequently validated20, providing the most convincing evidence for endocrine symptom burden when we designed TEXT and SOFT. An expanded module was used for centers with English as primary language to provide cross-validation between selected LASA indicators and the Center for Epidemiologic Studies-Depression Scale (CES-D)21 and the Medical Outcomes Study (MOS) sexual problems measures 22; these data will be presented elsewhere.

All indicators were in the linear analogue self-assessment (LASA) format.16 They were transformed to range from 0 to 100, with higher numbers reflecting a better condition. For all indicators, a clinically significant change was defined conservatively as at least ±8 points.23

We prospectively defined the following hypotheses:

  1. Primary
    1. There will be no difference in frequency of hot flushes, as measured by LASA format, between patients receiving exemestane+OFS versus tamoxifen+OFS in the short, intermediate and long-term. We expected an improvement of hot flushes at 24 months compared to six months.
    2. In patients without chemotherapy, there will be a differential effect on specific menopausal symptoms, i.e., more vaginal discharge in patients receiving tamoxifen+OFS, more vaginal dryness and more bone or joint pain in patients receiving exemestane+OFS.
    3. In patients without chemotherapy, there will be no difference in loss of sexual interest between those receiving exemestane+OFS versus tamoxifen+OFS.
  2. Secondary
    1. There will be no difference in global QoL domains (physical well-being, mood, coping effort) between patients receiving exemestane+OFS versus tamoxifen+OFS.
    2. Patients in TEXT receiving chemotherapy after randomization will report worse menopausal symptoms (primary endpoint: hot flushes), sexual functioning (primary endpoint: sexual interest) and global QoL domains (primary endpoint: coping effort) compared to those without chemotherapy at six month after randomization, and worse menopausal symptoms and sexual functioning at 60 months. Patients in SOFT with prior chemotherapy will have worse baseline scores and thus less worsening during endocrine therapy compared to those with concurrent chemotherapy in TEXT.
    3. Patients who report worse menopausal symptoms (primary endpoint: hot flushes) during the first 6 months on endocrine therapy will also report delayed adaptation in global QoL domains both during and beyond that time.

Outcomes

For each of the global and symptom-specific indicators in the QoL assessment, the changes from baseline to each timepoint were calculated, as timepoint score minus baseline score. The changes were summarized as mean with 95% confidence interval at each timepoint.

Statistical Analysis

The following patients were excluded from the QoL analysis: (1) patients with a QoL eligibility exemption; (2) patients at participating centers with poor overall QoL submission compliance (<60% of QoL assessments completed between baseline and 24 months; 7% of centers); (3) patients who otherwise had no QoL data submitted (Fig. 1).

Figure 1
CONSORT Flowchart to identify the 4096-patient intention-to-treat (ITT) quality-of-life population.

Mixed-effects linear modeling for repeated measures was used to test the short- (month 6), intermediate- (month 24) and long-term (month 60) effects of treatment on changes from baseline in each QoL indicator. The models included cohort (trial and chemotherapy use), treatment assignment (exemestane+OFS vs. tamoxifen+OFS), QL assessment timepoint, and the interactions of the three variables, without regard for statistical significance of the interaction parameters. Models were adjusted for baseline covariates: age, race/ethnicity, BMI, menstruation status, family history of breast or ovarian cancer, nodal status, tumor size, tumor grade, and HER2 status. An unstructured covariance was selected based on comparison of Akaike information criterion (AIC) and likelihood ratio tests. Within each cohort and overall, the estimated difference between treatment groups, 95% confidence intervals, and Wald P-values, were summarized at 6, 24 and 60 months. To address the primary hypotheses and secondary hypothesis 1, model contrasts compared treatment groups overall and separately according to cohort at the three specified timepoints. Tests of the cohort-by-treatment interaction were also performed.

To address secondary hypothesis 2, the difference between chemotherapy versus no chemotherapy was tested. To address secondary hypotheses 3, we added a new covariate, severity of menopausal symptoms during the first 6 months of therapy, by categorizing the patients into two groups (more vs. less severe) using the median value of the change in hot flushes from baseline to 6 months overall. The mean difference between the two severity groups in QoL change from baseline to 6, 12, 18, and 24 months was estimated and tested. The statistical analysis plan changed the approach to longitudinal modeling, rather than independent two-sample tests at each timepoint as specified the original protocols, because follow-up at the time of analysis was much longer than anticipated in the protocols and the modeling incorporated correlation of measurements over time and the multiple cohorts (defined by trial and chemotherapy use) analyzed.

To examine potential bias introduced by excluding patients at centers with poor QoL submission compliance, characteristics of those patients were summarized (Supplemental Table S3) and the analysis of 9 global and symptom-specific indicators specified in the hypotheses were re-run using all patients.

Although the hypotheses are stated as one-sided, tests were two-sided, consistent with the protocol. The reported P-values were not adjusted for testing of multiple QoL indicators over time, in order to look for consistency of the signal among comparable QL indicators. Power was considered within each trial protocol, but not re-estimated for the combined analysis. To have 80% power to detect an 8-unit difference between two treatment groups in the mean short-term change in hot flushes score (i.e., from baseline to 6 months), and in the mean intermediate-term change (baseline to 24 months), a sample size of 950 patients was estimated; for the indicator loss of sexual interest, approximately 1250 patients were required. The analysis used SAS version 9.4 (SAS Institute Inc., Cary, NC). The trials are registered with clinicaltrials.gov NCT00066703 (TEXT) and NCT00066690 (SOFT).

Role of the funding source

IBCSG was responsible for the TEXT and SOFT trial designs, including QoL, and for data collection and analysis. Pfizer and Ipsen, the manufacturers of exemestane and triptorelin, provided drug and Pfizer provided financial support; neither imposed restrictions on the investigators with respect to trial design or data collection, analysis or interpretation. The manuscript was prepared entirely by the authors. The Steering Committee (including employees of Pfizer and Ipsen) reviewed and approved the manuscript content. The statisticians (WL, MMR) and corresponding author (JB) had full access to all the data and the corresponding author had final responsibility to submit for publication.

RESULTS

Study Population

Between 7 November 2003 and 7 April 2011, 4717 premenopausal women were randomized to exemestane+OFS or tamoxifen+OFS. After exclusions (27 patients excluded from ITT efficacy analysis; 42 patients with a QoL eligibility exemption; 546 patients at participating centers with poor QoL submission compliance; 6 patients who had no QoL data submitted), 4096 remained in the intent-to-treat (ITT) QoL population (Fig. 1). The median follow-up was 5·7 years (IQR, 3·7 to 6·9 years).

The median age at randomization was 43 years and 42% (1726/4096) had node-positive disease (Table 1, Table S1). In TEXT, 59% (1400/2381) received concurrent chemotherapy, in SOFT, 55% (951/1715) received prior chemotherapy. Patient and disease characteristics at randomization were balanced across the treatment groups (Table S2). Patient, disease and treatment characteristics were generally balanced between the ITT QoL population and those excluded, though excluded patients had lower rates of physician-reported symptoms at randomization (Supplemental Table S3).

Table 1
Characteristics of patients in the TEXT and SOFT quality-of-life population, overall and according to patient cohort as defined by trial and chemotherapy stratum.1

The QoL form submission rates for the QoL population was similar in both trials, with 88% (14915/16856) submission of forms expected between baseline and 60 months in TEXT and 85% (10274/12037) in SOFT. The rates declined over time, with 90% (3606/4016) at six months, 86% (3337/3871) at 24 months, and 79% (2019/2542) at 60 months, and were balanced by treatment arm.

Treatment comparisons

Baseline scores most impaired included tiredness, sleep disturbance, being irritable, mood, coping effort, and health perception (Table 2). Vasomotor symptoms showed the greatest early change from baseline, as shown for hot flushes by cohort (Fig. 2). Hot flushes thereafter improved continuously (overall mean change months 24 - 6: exemestane+OFS= 6; tamoxifen+OFS= 8; both p<0·0001), but without reaching baseline scores. Overall, patients receiving exemestane+OFS experienced less worsening of hot flushes at six months than those receiving tamoxifen+OFS (Δ=8, P<0·0001; primary hypothesis 1). This difference persisted over time, although it became smaller. Patients receiving exemestane+OFS also experienced less worsening of sweats than those receiving tamoxifen+OFS (Fig. 3; Fig. S1; overall, month six: Δ=7, month 24 Δ=4, month 60: Δ=5, each P≤0·0001).

Figure 2
Hot flush scores (mean with 95% confidence intervals (CI); higher score indicates better condition) over time according to treatment assignment and patient cohort: (A) Hot flush scores from baseline to 60 months, separately according to chemotherapy cohort ...
Figure 3
Change of quality-of-life (QoL) symptom indicator scores from baseline to 6, 24, and 60 months, according to treatment assignment, overall across patient cohorts. The dashed vertical line at ±8 indicates the minimal clinically-meaningful change ...
Table 2
Baseline quality-of-life scores1 in the TEXT and SOFT quality-of-life population, overall and according to patient cohorts as defined by trial and chemotherapy stratum.

Gynecological and sexual symptoms were notable over the whole treatment period (Fig. 3 and 4 A, B; Fig. S2, S5–S8). Patients receiving exemestane+OFS experienced a greater exacerbation in vaginal dryness (overall, month six: Δ=−7, month 24: Δ=−8, month 60: Δ=−7, each P<0·0001), and less vaginal discharge than patients receiving tamoxifen+OFS (overall, month six: Δ=5, month 24: Δ=6, month 60: Δ=5, each P<0·0001; primary hypothesis 2) though the changes in vaginal discharge were not clinically meaningful (<8 points) in both treatment groups. Patients receiving exemestane+OFS also indicated a greater decline in sexual interest than those with tamoxifen+OFS. This difference increased over time (overall, month six: Δ=−5, month 24: Δ=−7, month 60: Δ=−8, each P<0·0001; primary hypothesis 3). At baseline, 76% (1548/2033 and 1556/2045) of the patients reported being sexually active in both treatment groups, and these percentages declined at six months to 69% (1261/1819) for patients receiving exemestane+OFS and 74% (1348/1827) for those receiving tamoxifen+OFS, at 24 months 70% (1171/1682) and 75% (1284/1705), and at 60 months 62% (641/1030) and 71% (702/992), respectively. If sexually active, patients receiving exemestane+OFS experienced greater changes in difficulties becoming aroused than patients receiving tamoxifen+OFS (overall, month six: Δ=−5, month 24: Δ=−6, each P<0·0001; month 60: Δ=−4, P=0.0068). There was no treatment difference in vaginal itching/irritation.

Figure 4
Vaginal dryness (A), loss of sexual interest (B), bone or joint pain (C), and troubled by weight gain (D) (mean with 95% confidence intervals (CI); higher score indicates better condition) from baseline to 60 months, separately according to chemotherapy ...

Irrespective of chemotherapy, patients receiving exemestane+OFS experienced a substantially greater worsening in bone or joint pain than patients receiving tamoxifen+OFS (Fig. 3 and and4C;4C; overall, month six: Δ=−15, month 24: Δ=−7, month 60: Δ=−5, each P<0·0001; primary hypothesis 2).

Patients in each treatment group felt similarly troubled by weight gain over the whole treatment period (Fig. 4D) and reported significant sleep disturbance in the short- and mid-term (Fig. S3). There were also no differences between the randomized treatments in appetite, feeling sick (nausea/vomiting), tiredness, headaches, being irritable or feeling dizzy (Fig. 3).

In general, patients reported small changes in physical well-being, mood and health perception in short-, mid- and long-term (all <8 points), but indicated less effort to cope over time (Fig. 5). There was no treatment difference in these global domains between patients receiving exemestane+OFS compared to those receiving tamoxifen+OFS (secondary hypothesis 1). Changes in treatment burden were smaller than those of the key symptoms.

Figure 5
Change of quality-of-life (QoL) global domain scores from baseline to 6, 24, and 60 months, according to treatment assignment, overall across patient cohorts. The dashed vertical line at ±8 indicates the minimal clinically-meaningful change of ...

Impact of chemotherapy

Patients treated without chemotherapy indicated no difference in treatment burden between the randomized treatments. In patients who received chemotherapy, those receiving exemestane+OFS reported being more bothered by treatment in short-term than those receiving tamoxifen+OFS, especially patients with prior chemotherapy (Fig. S5–S8; SOFT; month six: Δ=−7, P=0.00061, month 24: Δ=−5, P=0.021).

Patients in SOFT who received prior chemotherapy were randomized on average 8 months from final surgery, substantially later than the other three cohorts (Table 1). In general, patients with prior chemotherapy reported worse baseline scores for symptoms, possibly caused by chemotherapy, and thus smaller changes thereafter compared to those without chemotherapy, as shown for hot flushes (Fig. 2; secondary hypothesis 2).

Patients in TEXT receiving chemotherapy initiated it concurrently with triptorelin, but initiated exemestane or tamoxifen on average 19 weeks after randomization, later than the other three cohorts. They indicated a much smaller difference in bone or joint pain between exemestane+OFS and tamoxifen+OFS over the first 6 months (Fig. 4C; Δ of changes=−9) compared to the other three cohorts (Δ ranged −17 to −15). To note, there were no intermediate- to long-term QoL differences by chemotherapy in TEXT or SOFT.

Adaptation

At baseline, patients in SOFT reported better coping scores (with prior chemotherapy: 69, without: 70) than those in TEXT (with planned chemotherapy 59, without: 64). After 12 months, these differences disappeared.

Overall, patients with more severe hot flushes during the first 6 months had less improvement in coping scores in short- to mid-term than those with less severe menopausal symptoms during the first six months. This was most expressed in TEXT patients receiving concurrent chemotherapy (month six: Δ=−10, P<0·0001; month 24: Δ=−6, P=0.00019; secondary hypothesis 3). For changes in mood and physical well-being, the difference between the two severity groups was consistent although smaller (data not shown).

DISCUSSION

In these premenopausal trials with OFS, the comparison of exemestane with tamoxifen showed differential QoL findings. Patients assigned exemestane+OFS reported significantly more detrimental effects of bone or joint pain, vaginal dryness, greater loss of sexual interest and difficulties becoming aroused, while patients assigned tamoxifen+OFS were significantly more affected by hot flushes, sweats and vaginal discharge. In general, patients reported considerable changes in key endocrine symptoms in the short-, mid- and long-term, but differences between the randomized treatments were small. The changes in global QoL domains were similar between the randomized treatment groups. The differential findings are in contrast to no differences (except for vaginal discharge) in IES in postmenopausal women.5

Vasomotor symptoms were most expressed over the whole treatment period, although they improved in the mid- and long-term. Contrary to our hypothesis, patients receiving tamoxifen+OFS were more affected by hot flushes than patients on exemestane+OFS.

The continuous impact of gynecological and sexual symptoms are a serious concern. Patients on exemestane+OFS reported more vaginal dryness, greater loss of sexual interest, more difficulties in becoming aroused, and were less likely to remain sexually active while patients on tamoxifen+OFS reported more but small changes in vaginal discharge. These differences persisted over time. Worsening in bone or joint pain was more expressed in patients with exemestane+OFS, particularly in the short-term, than in patients with tamoxifen+OFS.

Patients indicated a clinically-relevant change in their treatment burden mainly in the short-term. The changes in this summative measure were smaller than might be expected from the side-effect profile. An early difference in favor of tamoxifen+OFS in patients who received chemotherapy did not persist. Similarly, there were no differences between the randomized treatments in any of the other global domains, in agreement with previous trials.4,5,9 The relatively low treatment burden may reflect patients’ adaptation, as suggested by their decreasing coping effort. This interpretation is supported by preference studies.24,25 However, a substantial minority of patients is expected to suffer from burdensome endocrine symptoms,9 as reflected in the 14% of patients in the total sample who had stopped all protocol-assigned treatments early (16% exemestane+OFS and 11% tamoxifen+OFS), most reporting targeted adverse events.1 In general, the patient-reported symptoms corresponded to the reported adverse events. The patient-reported outcomes show the perceived time course of these symptoms and thus give a more comprehensive and accurate picture of the net impact of a given symptom (e.g., long-term impact on sexual function).

An early recognition of symptom burden is imperative for patient care. Some of the main menopausal symptoms can be ameliorated by a multidisciplinary approach.26,27 As shown for hot flushes, pronounced symptom burden had a relevant impact on patients’ adaptation.

Chemotherapy was optional in both trials, and therefore associated with prognostic factors. In the context of OFS for all patients, and thus in contrast to the ZIPP trial,10 the differential endocrine effects were similarly present in patients with or without chemotherapy. In TEXT, patients with or without concurrent chemotherapy required comparable effort to cope14 at month 6. This differs from the Zoladex Early Breast Cancer Research Association (ZEBRA) trial and IBCSG Trial VIII, both with OFS as a randomized intervention, in which patients receiving goserelin required substantially less effort to cope14 at month 6 than those receiving chemotherapy (CMF).28,29

Chemotherapy had a clinically-relevant short-term impact in that patients with prior chemotherapy (in SOFT) had lower baseline scores in chemotherapy-related domains and subsequently a greater improvement. However, these patients had more time to adapt than those with planned chemotherapy and reported less coping effort at baseline. A timing effect was obvious by the smaller difference between exemestane and tamoxifen in bone or joint pain in short-term changes in TEXT patients with planned chemotherapy. These patients received chemotherapy and triptorelin concurrently after randomization, and exemestane or tamoxifen several months later than the other three cohorts.

The inclusion of the four cohorts in the analysis adds complexity to the interpretation of the results. On the other hand, the use or not of chemotherapy is a decision for each patient; when chemotherapy is used, OFS will be initiated concurrently for some patients and may be initiated after chemotherapy when premenopausal status is established or re-established. The analysis provides insight into the patterns of QoL over time for each of these scenarios.

When we developed these trials at the beginning of the last decade, we were not aware of any endocrine symptom questionnaire ready to be used in the required languages. Due to feasibility reasons, specific QoL domains which are especially relevant in this clinical context (e.g., depression, body image, sexual function, cognitive function) could not be investigated in depth in these large-scale trials. Instead, we used a pragmatic measurement approach based on LASA indicators as previously used in IBCSG trials. Depression, sexual and cognitive functioning are under further investigation in a subset of TEXT and SOFT.

Longitudinal QoL data submission always presents a challenge. More than 500 centers world-wide participated in the trials and we excluded 7% of centers with poor QoL form submission rates. As well, the QoL form submission rate decreased in the long-term, but overall was balanced between the randomized treatments. Sensitivity analysis confirmed that results were robust to the analysis approach for missing data.

In conclusion, from a QoL perspective, there is no strong indication to favor either exemestane+OFS or tamoxifen+OFS, and the benefit in disease-free survival of exemestane+OFS over tamoxifen+OFS observed in TEXT and SOFT is achieved without a detrimental impact on global QoL. The differential effects of the two treatments on endocrine symptoms should be considered with each patient individually.

REASEARCH IN CONTEXT PANEL

Evidence before this study

We searched PubMed for clinical trials published in English between Jan 1, 1998, and Jan. 1, 2003, and updated it on February 26, 2015, that assessed patient-reported symptoms and QoL in premenopausal women with early breast cancer undergoing adjuvant endocrine treatment. We used the search terms “premenopausal”, “tamoxifen”, “aromatase inhibitors”, “QoL”, and “patient-reported outcome”. We manually reviewed the results and found no randomized phase III trial reporting on patient-reported symptoms and QoL on the specific comparison of tamoxifen with an aromatase inhibitor in premenopausal women. Existing data is based on data from postmenopausal women35 or placebo-controlled breast cancer prevention trials.69

Added value of this study

In TEXT and SOFT trials, patients assigned exemestane+OFS reported more detrimental effects of bone or joint pain, vaginal dryness, greater loss of sexual interest and difficulties becoming aroused, while patients assigned tamoxifen+OFS were more affected by hot flushes and sweats. Global QoL domains were similar between the randomized treatment groups. The differential impact on endocrine symptoms is only partly consistent with the findings of previous trials of an aromatase inhibitor compared with tamoxifen in postmenopausal patients (without OFS). In the IES, comparisons between exemestane and tamoxifen revealed no differences for any of the endocrine symptoms except for vaginal discharge, which was greater with tamoxifen up to 24 months following 2–3 years of treatment with tamoxifen.5 In the ATAC trial, self-reported gynecological symptoms were more frequent with anastrozole compared to tamoxifen, whereas dizziness, cold sweats and vaginal discharge were more frequent with tamoxifen.3,4 Patient-reported bone or joint pain was not assessed in these trials.

Implication of all the available evidence

In premenopausal patients with hormone receptor-positive breast cancer, the TEXT and SOFT trials combined analysis showed a disease-free survival benefit for exemestane plus ovarian function suppression (OFS) compared with tamoxifen+OFS, providing a new treatment option for premenopausal women who receive OFS as part of adjuvant endocrine therapy.1 From a QoL perspective, there is no strong indication to favor either exemestane+OFS or tamoxifen+OFS. The differential effects of the two treatments on endocrine symptoms should be considered with each patient individually.

Supplementary Material

ACKNOWLEDGMENT

We thank the patients, physicians, nurses, and trial coordinators who participated in the TEXT and SOFT clinical trials. TEXT and SOFT receive financial support for trial conduct from Pfizer, the International Breast Cancer Study Group and the US National Cancer Institute. Pfizer and Ipsen provide drug supply. Support for the coordinating group, IBCSG: Frontier Science and Technology Research Foundation, Swiss Group for Clinical Cancer Research (SAKK), US National Cancer Institute (NCI) (CA75362 to RG, MMR), Cancer Research Switzerland/Oncosuisse, and the Foundation for Clinical Cancer Research of Eastern Switzerland (OSKK). Grant support of cooperative groups: Australia and New Zealand Breast Cancer Trials Group (NHMRC 351161 and 510788); SWOG (US NIH CA32102); Alliance (US NIH CA180821); ECOG-ACRIN (US NIH CA21115 and CA16116); NSABP/NRG (US NIH U10-CA-12027, U10-CA-69651, U10-CA-37377, U10-CA-69974); NCIC (US NIH CA077202 and CCSRI 015469 and 021039).

M.C. reports personal fees from Taiho Pharmaceutical, personal fees from Novartis, personal fees from AstraZeneca, personal fees from Pierre Fabre, personal fees from AbbVie, outside the submitted work; M.M.R. reports grants from Pfizer, Ipsen, during the conduct of the study; R.D.G. reports grants from Pfizer, during the conduct of the study; grants from AstraZeneca, grants from Novartis, grants from Celgene, grants from Merck, grants from Roche, grants from GSK, outside the submitted work.

Footnotes

Clinicaltrials.gov NCT00066703 (TEXT) and NCT00066690 (SOFT)

DISCLOSURES

All other authors declared no conflicts of interest.

AUTHOR CONTRIBUTIONS

J.B., K.R., O.P., P.A.F., G.F.F., B.A.W., A.S.C., A.G., and R.D.G. participated in the design.

M.C., H.J.B., C.T., G.P., S.S., T.R., F.P., L.P., V.P., O.P., G.F.F., P.A.F., A.G., B.A. W. participated in data collection. W.L., M.M.R., J.B., K.R., K.N.P., R.D.G. participated in data analysis. All authors participated in data interpretations, drafting and finalising the report.

Contributor Information

Jürg Bernhard, International Breast Cancer Study Group Coordinating Center and Bern University Hospital, Inselspital, Bern, Switzerland.

Weixiu Luo, International Breast Cancer Study Group Statistical Center, Dana-Farber Cancer Institute, Boston, MA, USA.

Karin Ribi, International Breast Cancer Study Group Coordinating Center, Bern, Switzerland.

Marco Colleoni, Division of Medical Senology, European Institute of Oncology and International Breast Cancer Study Group, Milan, Italy.

Harold J. Burstein, Dana-Farber Cancer Institute and Alliance for Clinical Trials in Oncology, Boston, MA, USA.

Carlo Tondini, Medical Oncology, Osp. Papa Giovanni XXIII and International Breast Cancer Study Group, Bergamo, Italy.

Graziella Pinotti, Medical Oncology, Ospedale di Circolo and Fondazione Macchi and International Breast Cancer Study Group, Varese, Italy.

Simon Spazzapan, Medical Oncology, Centro di Riferimento Oncologico and International Breast Cancer Study Group, Aviano, Italy.

Thomas Ruhstaller, Breast Center, Kantonsspital St. Gallen, Swiss Group for Clinical Cancer Research (SAKK) and International Breast Cancer Study Group, St. Gallen, Switzerland.

Fabio Puglisi, Department of Oncology, University Hospital of Udine, University of Udine and International Breast Cancer Study Group, Udine, Italy.

Lorenzo Pavesi, Oncology Unit, Salvatore Maugeri Foundation and International Breast Cancer Study Group, Pavia, Italy.

Vani Parmar, Surgical Oncology, Breast Services, Tata Memorial Hospital and International Breast Cancer Study Group, Mumbai, India.

Meredith M. Regan, International Breast Cancer Study Group Statistical Center, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA, USA.

Olivia Pagani, Institute of Oncology of Southern Switzerland, Swiss Group for Clinical Cancer Research (SAKK), and International Breast Cancer Study Group, Lugano Viganello, Switzerland.

Gini F. Fleming, The University of Chicago Medical Center and Alliance for Clinical Trials in Oncology, Chicago, IL, USA.

Prudence A. Francis, Peter MacCallum Cancer Center, St Vincent’s Hospital, University of Melbourne, Melbourne and Australia & New Zealand Breast Cancer Trials Group, Newcastle, Australia; International Breast Cancer Study Group.

Karen N. Price, International Breast Cancer Study Group Statistical Center, Frontier Science and Technology Research Foundation, Boston, MA, USA.

Alan S. Coates, International Breast Cancer Study Group and University of Sydney, Sydney, Australia.

Richard D. Gelber, International Breast Cancer Study Group Statistical Center, Dana-Farber Cancer Institute, Harvard Medical School, Harvard School of Public Health, Frontier Science and Technology Research Foundation, Boston, MA, USA.

Aron Goldhirsch, Program for Breast Health (Senology), European Institute of Oncology, Milan, Italy & International Breast Cancer Study Group, Bern, Switzerland.

Barbara A. Walley, Tom Baker Cancer Centre and NCIC Clinical Trials Group, Calgary, AB, Canada.

REFERENCES

1. Pagani O, Regan MM, Walley BA, et al. Adjuvant exemestane with ovarian suppression in premenopausal breast cancer. The New England journal of medicine. 2014;371(2):107–118. [PMC free article] [PubMed]
2. Regan MM, Pagani O, Fleming GF, et al. Adjuvant treatment of premenopausal women with endocrine-responsive early breast cancer: design of the TEXT and SOFT trials. Breast. 2013;22(6):1094–1100. [PMC free article] [PubMed]
3. Fallowfield L, Cella D, Cuzick J, Francis S, Locker G, Howell A. Quality of life of postmenopausal women in the Arimidex, Tamoxifen, Alone or in Combination (ATAC) Adjuvant Breast Cancer Trial. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2004;22(21):4261–4271. [PubMed]
4. Cella D, Fallowfield L, Barker P, et al. Quality of life of postmenopausal women in the ATAC ("Arimidex", tamoxifen, alone or in combination) trial after completion of 5 years' adjuvant treatment for early breast cancer. Breast cancer research and treatment. 2006;100(3):273–284. [PubMed]
5. Fallowfield LJ, Bliss JM, Porter LS, et al. Quality of life in the intergroup exemestane study: a randomized trial of exemestane versus continued tamoxifen after 2 to 3 years of tamoxifen in postmenopausal women with primary breast cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2006;24(6):910–917. [PubMed]
6. Day R, Ganz PA, Costantino JP, Cronin WM, Wickerham DL, Fisher B. Health-related quality of life and tamoxifen in breast cancer prevention: a report from the National Surgical Adjuvant Breast and Bowel Project P-1 Study. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 1999;17(9):2659–2669. [PubMed]
7. Fallowfield L, Fleissig A, Edwards R, et al. Tamoxifen for the prevention of breast cancer: psychosocial impact on women participating in two randomized controlled trials. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2001;19(7):1885–1892. [PubMed]
8. Goss PE, Ingle JN, Ales-Martinez JE, et al. Exemestane for breast-cancer prevention in postmenopausal women. The New England journal of medicine. 2011;364(25):2381–2391. [PubMed]
9. Maunsell E, Goss PE, Chlebowski RT, et al. Quality of life in MAP.3 (Mammary Prevention 3): a randomized, placebo-controlled trial evaluating exemestane for prevention of breast cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2014;32(14):1427–1436. [PMC free article] [PubMed]
10. Nystedt M, Berglund G, Bolund C, Fornander T, Rutqvist LE. Side effects of adjuvant endocrine treatment in premenopausal breast cancer patients: a prospective randomized study. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2003;21(9):1836–1844. [PubMed]
11. Bernhard J, Hurny C, Coates AS, et al. Quality of life assessment in patients receiving adjuvant therapy for breast cancer: the IBCSG approach. The International Breast Cancer Study Group. Annals of oncology : official journal of the European Society for Medical Oncology / ESMO. 1997;8(9):825–835. [PubMed]
12. Butow P, Coates A, Dunn S, Bernhard J, Hurny C. On the receiving end. IV: Validation of quality of life indicators. Annals of oncology : official journal of the European Society for Medical Oncology / ESMO. 1991;2(8):597–603. [PubMed]
13. Hurny C, Bernhard J, Coates A, et al. Responsiveness of a single-item indicator versus a multi-item scale: assessment of emotional well-being in an international adjuvant breast cancer trial. Medical care. 1996;34(3):234–248. [PubMed]
14. Hurny C, Bernhard J, Bacchi M, et al. The Perceived Adjustment to Chronic Illness Scale (PACIS): a global indicator of coping for operable breast cancer patients in clinical trials. Swiss Group for Clinical Cancer Research (SAKK) and the International Breast Cancer Study Group (IBCSG) Support Care Cancer. 1993;1(4):200–208. [PubMed]
15. Hurny C, van Wegberg B, Bacchi M, et al. Subjective health estimations (SHE) in patients with advanced breast cancer: an adapted utility concept for clinical trials. British journal of cancer. 1998;77(6):985–991. [PMC free article] [PubMed]
16. Bernhard J, Sullivan M, Hurny C, Coates AS, Rudenstam CM. Clinical relevance of single item quality of life indicators in cancer clinical trials. British journal of cancer. 2001;84(9):1156–1165. [PMC free article] [PubMed]
17. Bernhard J, Maibach R, Thurlimann B, Sessa C, Aapro MS. Swiss Group for Clinical Cancer R. Patients' estimation of overall treatment burden: why not ask the obvious? Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2002;20(1):65–72. [PubMed]
18. Ganz PA, Day R, Ware JE, Jr, Redmond C, Fisher B. Base-line quality-of-life assessment in the National Surgical Adjuvant Breast and Bowel Project Breast Cancer Prevention Trial. Journal of the National Cancer Institute. 1995;87(18):1372–1382. [PubMed]
19. Greendale GA, Reboussin BA, Hogan P, et al. Symptom relief and side effects of postmenopausal hormones: results from the Postmenopausal Estrogen/Progestin Interventions Trial. Obstetrics and gynecology. 1998;92(6):982–988. [PubMed]
20. Alfano CM, McGregor BA, Kuniyuki A, et al. Psychometric properties of a tool for measuring hormone-related symptoms in breast cancer survivors. Psycho-oncology. 2006;15(11):985–1000. [PMC free article] [PubMed]
21. Radloff LS. The CES-D Scale: a self-reported depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401.
22. Sherbourne CD. Social functioning: sexual problems measures. In: Stewart AL, Ware JE, editors. Measuring functioning and well-being: the Medical Outcomes Study approach. Durham NC: Duke Univ Press; 1992. pp. 194–204.
23. Sloan JA, Dueck A. Issues for statisticians in conducting analyses and translating results for quality of life end points in clinical trials. Journal of biopharmaceutical statistics. 2004;14(1):73–96. [PubMed]
24. Duric VM, Fallowfield LJ, Saunders C, Houghton J, Coates AS, Stockler MR. Patients' preferences for adjuvant endocrine therapy in early breast cancer: what makes it worthwhile? British journal of cancer. 2005;93(12):1319–1323. [PMC free article] [PubMed]
25. Thewes B, Meiser B, Duric VM, et al. What survival benefits do premenopausal patients with early breast cancer need to make endocrine therapy worthwhile? The Lancet Oncology. 2005;6(8):581–588. [PubMed]
26. Hickey M, Saunders CM, Stuckey BG. Management of menopausal symptoms in patients with breast cancer: an evidence-based approach. The Lancet Oncology. 2005;6(9):687–695. [PubMed]
27. Drewe J, Bucher KA, Zahner C. A systematic review of non-hormonal treatments of vasomotor symptoms in climacteric and cancer patients. Springer Plus. 2015;4:65. [PMC free article] [PubMed]
28. de Haes H, Olschewski M, Kaufmann M, et al. Quality of life in goserelin-treated versus cyclophosphamide + methotrexate + fluorouracil-treated premenopausal and perimenopausal patients with node-positive, early breast cancer: the Zoladex Early Breast Cancer Research Association Trialists Group. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2003;21(24):4510–4516. [PubMed]
29. Bernhard J, Zahrieh D, Castiglione-Gertsch M, et al. Adjuvant chemotherapy followed by goserelin compared with either modality alone: the impact on amenorrhea, hot flashes, and quality of life in premenopausal patients--the International Breast Cancer Study Group Trial VIII. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2007;25(3):263–270. [PubMed]