Abstract
Background:
TAILORx prospectively assessed fatigue and endocrine symptoms among women with early-stage hormone receptor-positive breast cancer, and mid-range risk of recurrence, randomized to endocrine (E) versus chemoendocrine therapy (CT+E).
METHODS:
Participants completed the Functional Assessment of Chronic Illness Therapy Fatigue, Patient Reported Outcomes Measurement Information System Fatigue Short Form, and Functional Assessment of Cancer Therapy Endocrine Symptoms at baseline, 3, 6, 12, 24, and 36 months. We used linear regression to model outcomes on baseline symptoms, treatment and other factors.
RESULTS:
Participants (n=458) in both treatment arms reported greater fatigue and endocrine symptoms at early follow-up compared to baseline. The magnitude of change in fatigue was significantly greater for CT+E than E at 3 and 6, but not at 12, 24, and 36 months. The CT+E arm reported significantly greater changes in endocrine symptoms from baseline to 3 months compared to E; change scores were not significantly different at later timepoints. Endocrine symptom trajectories by treatment were different by menopausal status, with the effect larger and increasing for post-menopausal patients.
CONCLUSIONS:
Adjuvant CT+E was associated with greater increases in fatigue and endocrine symptoms at early timepoints compared to E. These differences lessened over time, demonstrating early more than long-term chemotherapy effects. Treatment arm differences in endocrine symptoms were more evident in post-menopausal patients.
Keywords: breast neoplasms, patient-reported outcomes, drug therapy, fatigue, hormones
Lay summary:
TAILORx participants with early-stage hormone receptor-positive breast cancer, and intermediate risk of recurrence, were randomly assigned to endocrine (E) versus chemoendocrine therapy (CT+E). 458 women reported fatigue and endocrine symptoms at baseline, 3, 6, 12, 24, and 36 months. Both groups reported greater symptoms at early follow-up compared to baseline. Increases in fatigue were greater for CT+E than E at three and six months but not later. The CT+E group reported greater changes in endocrine symptoms compared to the E group at three months but not later.
Precis:
TAILORx participants randomized to chemoendocrine therapy reported significant increases in fatigue and endocrine symptoms three months after randomization, which decreased through 36 months. Those randomized to endocrine therapy also reported increased symptoms from baseline, although of lesser magnitude.
Breast cancer remains the most common cancer in women1 but mortality rates are declining, partially due to widespread adjuvant therapy2. Hormone receptor-positive (HR+) cases account for roughly two thirds of breast cancers, and can be treated with chemotherapy and/or endocrine therapy3. Although less toxic than chemotherapy, long-term endocrine therapy4,5 can produce symptoms (arthralgias, vasomotor symptoms, sexual dysfunction) that impact health-related quality of life (HRQoL)6 and medication non-adherence7,8, which in turn decreases treatment efficacy9.
Beyond being acutely toxic, chemotherapy may also result in future health consequences, including persistent fatigue10. Following the Early Breast Cancer Trialists Group meta-analysis11, adjuvant chemotherapy became standard for most localized breast cancers12,13. The ensuing “over-treatment,” led to development of the 21-gene assay4,14,15 to predict: risk of distant recurrence more accurately than classical clinicopathologic features in patients with HR+ breast cancer14,16, and benefit of adding adjuvant chemotherapy to endocrine therapy in that population5,7.
The Trial Assigning Individualized Options for Treatment (TAILORx), randomized women with HR+, human epidermal growth factor receptor 2 (HER2)-negative, axillary node-negative early breast cancer and intermediate recurrence scores (RSs=11–25) to chemotherapy followed by endocrine therapy (CT+E) versus endocrine therapy (E).17,18. TAILORx demonstrated highly favorable outcomes for patients with RSs 0–25 receiving endocrine therapy, indicating that women with intermediate or low RSs, can be spared chemotherapy.16,19 TAILORx provided an unparalleled opportunity to prospectively evaluate the trajectory of HRQoL among women randomized to CT+E versus E for breast cancer.
Patient-reported outcome (PRO) measures are ideal for assessing subjective symptoms, and can inform treatment20,21. TAILORx allowed us to examine the unique contributions of chemotherapy to fatigue22 and endocrine symptoms6. The substudy’s primary objective was to compare those longitudinal patient-reported symptoms among women with early HR+ breast cancer randomized to adjuvant CT+E versus E.
METHODS
Design & Participants
The ECOG-ACRIN Cancer Research Group coordinated TAILORx (ClinicalTrials.gov: NCT00310180)17,18, which enrolled patients from 4/2006–10/2010. In January 2010, a PRO substudy was approved by participating institutions’ human investigations committees. Eligibility was consistent with TAILORx17,18: women 18–75 years old, diagnosed with HR+, HER2-negative, axillary node-negative breast cancer, meeting guidelines for adjuvant chemotherapy. Participants provided informed consent and completed PRO measures at baseline (pre-randomization) and 3, 6, 12, 24 and 36 months afterward. Given menopausal status was among the TAILORx stratification factors, we conducted subset analyses to examine its relationships to fatigue and endocrine symptoms. We retrieved data in February, 2016.
Measures
The Patient Reported Outcomes Measurement Information System Fatigue Short Form (PROMIS Fatigue 7) is a seven-item measure of fatigue using a 5-point Likert response scale23. PROMIS measures are reported on a T score metric, with higher scores indicating greater symptomatology. It has demonstrated good precision and reliability24. Here, Cronbach’s alpha was 0.874.
The Functional Assessment of Chronic Illness Therapy Fatigue (FACIT Fatigue) Scale is a 13-item measure using a 5-point Likert scale25, with higher scores indicating less fatigue (i.e., higher HRQoL). It has demonstrated reliability and validity in clinical trials. Cronbach’s alpha was 0.956.
The Functional Assessment of Cancer Therapy Endocrine Symptoms (FACT-ES) is a 19-item measure using a 5-point Likert response scale, with higher scores indicating less symptomatology. It has demonstrated suitability for clinical trials26. Cronbach’s alpha was 0.844. Trial participants completed additional PROs (Functional Assessment of Cancer Therapy–Cognitive Function [FACT-Cog]27,28, FACT-General29 & Assessment of Survivor Concerns)30, with results reported separately31.
Analysis
Our primary analysis compared TAILORx participants who received the treatments to which they were randomized. These per-protocol analyses excluded patients based on post-randomization treatment decisions, which may introduce bias. To examine their robustnes, we also performed intention-to-treat (ITT) analyses. Our primary endpoints were fatigue and endocrine symptom score differences between treatment arms at 3 months, controlling for baseline scores. Most women on CT+E received 12-week regimens and were expected to experience maximum chemotherapy side effects at 3 months. We also examined treatment differences at subsequent timepoints. The PRO substudy was designed to have 90% power for a 4.5 point difference in mean change from baseline to 3 months in FACT-Cog (primary PRO endpoint) between CT+E and E31; a sample of 235 women per arm was required. We expected comparable power for differences of similar magnitude in the three PRO measures examined here.
We computed means and SDs using all cases at a timepoint. When comparing treatment arms at timepoints, the analysis fits a linear model with arm (binary covariate) and baseline levels (continuous linear covariate), with the test and estimated effect based on the coefficient of treatment effect. We also computed mean changes from baseline and standard errors. We included only cases with assessments at baseline and follow-up timepoint. We excluded cases with baseline assessments >7 days after treatment initiation. We conducted analyses with R 3.5.132.
RESULTS
Sample
734 women enrolled in the PRO substudy (Figure 1). They had generally similar characteristics to the larger TAILORx31. Participants were eligible to continue on PRO substudy if they experienced a recurrence or new primary breast cancer while enrolled on TAILORx. Table 1 gives the demographic and clinical characteristics for patients (n= 458) in the per-protocol analysis with data on at least one of the PRO measures at baseline and 3 months. As previously reported, the characteristics of participants in the PRO study, as compared to the larger trial sample randomized to treatment, were generally very similar, with a slightly higher proportion of postmenopausal patients, low recurrence score and very small (< 1cm) tumors in the PRO study.31 The per-protocol dataset had slightly lower proportions of older patients and lower RSs, low-grade and very small tumors on the CT+E arm (such patients may have been more likely to refuse chemotherapy).
Figure 1.
CONSORT diagram – TAILORx PRO substudy
PRO= patient-reported outcome; CT+E= chemotherapy followed by endocrine therapy; E= endocrine therapy alone;
1Reasons for missing baseline data: FACT ES= patient not given PRO form (n=15), refusal (n=15), language or disability (n=14), other (n=8), site did not provide reason (n=11); FACIT-Fatigue= patient not given PRO form (n=13), refusal (n=l5), language or disability (n=14), other (n=8), site did not provide reason (n=12); PROMIS-Fatigue: patient not given PRO form (n=17), refusal (n=16), language or disability (n=14), other (n=9), site did not provide reason (n=11).
2Numbers presented in order FACT-ES, FACIT-Fatigue, PROMIS Fatigue.
*Characteristics of this sample of 458 per protocol participants, the largest with data on at least one of the PRO measures at baseline and 3 months, are presented in Table 1.
Table 1.
Demographic & Clinical Characteristics (n=458*)
| E(n=238) | CT+E (n=220) | |
|---|---|---|
| Mean age (SD) years | 56 (9) | 55 (8) |
| Age | ||
| <50 | 78 (33%) | 68 (31%) |
| 51–65 | 115 (48%) | 126 (57%) |
| >65 | 45 (19%) | 26 (12%) |
| Race | ||
| White | 196 (82%) | 181 (82%) |
| Black | 15 (6%) | 13 (6%) |
| Asian | 16 (7%) | 8 (4%) |
| Other/Unknown | 11 (5%) | 18 (8%) |
| Ethnicity | ||
| Hispanic | 12 (5%) | 18 (8%) |
| Non-Hispanic | 210 (88%) | 183 (83%) |
| Unknown | 16 (7%) | 19 (9%) |
| Menopause | ||
| Pre | 74 (31%) | 80 (36%) |
| Post | 164 (69%) | 140 (64%) |
| Recurrence score | ||
| 11–15 | 101 (42%) | 82 (37%) |
| 16–20 | 81 (34%) | 80 (36%) |
| 21–25 | 56 (24%) | 58 (26%) |
| Tumor size, cm | ||
| <=1.0 | 37 (16%) | 21 (10%) |
| 1.1–2.0 | 149 (63%) | 140 (64%) |
| 2.1–3.0 | 40 (17%) | 50 (23%) |
| 3.1–4.0 | 11 (5%) | 5 (2%) |
| >4.0 | 1 (0%) | 4 (2%) |
| Unknown | 0 | 0 |
| Histology grade | ||
| Low | 75 (32%) | 59 (27%) |
| Medium | 123 (53%) | 127 (58%) |
| High | 33 (14%) | 32 (15%) |
| Unknown | 7 | 2 |
| Estrogen receptor | ||
| Negative | 0 (0%) | 0 (0%) |
| Positive | 238 (100%) | 220 (100%) |
| Progesterone receptor | ||
| Negative | 14 (6%) | 18 (8%) |
| Positive | 216 (94%) | 195 (92%) |
| Unknown | 8 | 7 |
| Surgery | ||
| Tumorectomy | 175 (74%) | 152 (69%) |
| Mastectomy | 63 (26%) | 68 (31%) |
| Initial endocrine therapy | ||
| Aromatase inhibitor | 139 (58%) | 127 (58%) |
| Tamoxifen | 87 (37%) | 83 (38%) |
| Tamoxifen & Aromatase inhibitor | 3 (1%) | 5 (2%) |
| Ovarian function suppression | 7 (3%) | 0 (0%) |
| Other | 1 (0%) | 0 (0%) |
| None reported | 1 (0%) | 5 (2%) |
| Changed endocrine therapy | ||
| Tamoxifen to Aromatase inhibitor | 31 (13%) | 41 (19%) |
| Aromatase inhibitor to Tamoxifen | 14 (6%) | 21 (10%) |
| Chemotherapy | ||
| Taxane & cyclophosphamide | -- | 153 (70%) |
| Anthracycline +/− taxane | -- | 44 (20%) |
| CMF | -- | 18 (8%) |
| Other | -- | 5 (2%) |
| None | 238 (100%) | 0 (0%) |
| Comorbidities | ||
| Hypertension: No | 140 (59%) | 134 (62%) |
| Yes | 97 (41%) | 83 (38%) |
| Unknown | 1 | 3 |
| Hyperlipidemia: No | 165 (70%) | 164 (76%) |
| Yes | 71 (30%) | 51 (24%) |
| Unknown | 2 | 5 |
| Depression: No | 186 (79%) | 172 (80%) |
| Yes | 50 (21%) | 43 (20%) |
| Unknown | 2 | 5 |
| Diabetes: No | 210 (89%) | 192 (89%) |
| Yes | 25 (11%) | 23 (11%) |
| Unknown | 3 | 5 |
| Osteoarthritis: No | 214 (90%) | 191 (88%) |
| Yes | 23 (10%) | 25 (12%) |
| Unknown | 1 | 4 |
| Osteopenia/Osteoporosis: No | 199 (85%) | 201 (94%) |
| Yes | 36 (15%) | 13 (6%) |
| Unknown | 3 | 6 |
• The per-protocol analytic data set was specified a priori and consists of patients (n= 458) with data on at least one of the patient-reported outcome measures at baseline and 3 months
Treatment Arm-related Differences
In the per-protocol analysis, women in CT+E reported significantly greater increases in PROMIS Fatigue 7 scores from baseline to 3 and 6 months compared to women in E; change scores were comparable between treatment arms at at later timepoints (Table 2). The trajectories of longitudinal PROMIS Fatigue change scores by treatment arm converged more over time because the CT+E arm reported decreased fatigue after a sharp increase post-baseline (while receiving chemotherapy). However, both arms reported more fatigue at all follow-up assessments compared to baseline (Figure 2a).
Table 2.
Per Protocol Analysis - Changes from Baseline
| Subset | Timepoint | n | E | CT+E | Raw Diff | LM Diff | p.LM |
|---|---|---|---|---|---|---|---|
| FACT-ES | |||||||
| All | 3-month | 458 | −3.61 (0.59) | −5.56 (0.60) | −1.95 (0.84) | −1.62 (0.79) | 0.04 |
| All | 6-month | 467 | −4.24 (0.60) | −5.63 (0.55) | −1.39 (0.81) | −0.97 (0.76) | 0.20 |
| All | 12-month | 451 | −5.62 (0.67) | −6.96 (0.68) | −1.34 (0.95) | −1.08 (0.90) | 0.23 |
| All | 24-month | 385 | −5.31 (0.75) | −6.81 (0.68) | −1.50 (1.02) | −1.05 (0.96) | 0.27 |
| All | 36-month | 337 | −5.17 (0.80) | −7.14 (0.85) | −1.97 (1.17) | −1.69 (1.10) | 0.13 |
| Pre-Menopausal | 3-month | 154 | −5.96 (1.14) | −7.62 (1.02) | −1.65 (1.53) | −1.44 (1.47) | 0.33 |
| Pre-Menopausal | 6-month | 151 | −6.19 (1.15) | −8.34 (1.03) | −2.15 (1.54) | −1.63 (1.45) | 0.26 |
| Pre-Menopausal | 12-month | 148 | −8.95 (1.16) | −7.94 (1.28) | 1.01 (1.73) | 1.06 (1.64) | 0.52 |
| Pre-Menopausal | 24-month | 118 | −10.39 (1.53) | −8.29 (1.27) | 2.09 (1.99) | 2.27 (1.84) | 0.22 |
| Pre-Menopausal | 36-month | 102 | −10.84 (1.70) | −8.96 (1.66) | 1.88 (2.38) | 2.18 (2.25) | 0.34 |
| Pre-Menopausal | 3-month | 304 | −2.55 (0.66) | −4.39 (0.72) | −1.83 (0.98) | −1.49 (0.92) | 0.11 |
| Post-Menopausal | 6-month | 316 | −3.41 (0.69) | −4.19 (0.61) | −0.78 (0.93) | −0.45 (0.87) | 0.60 |
| Post-Menopausal | 12-month | 303 | −4.10 (0.79) | −6.45 (0.78) | −2.34 (1.12) | −2.04 (1.06) | 0.06 |
| Post-Menopausal | 24-month | 267 | −3.23 (0.80) | −6.10 (0.80) | −2.87 (1.13) | −2.39 (1.06) | 0.03 |
| Post-Menopausal | 36-month | 235 | −2.87 (0.82) | −6.28 (0.97) | −3.41 (1.26) | −3.17 (1.18) | 0.008 |
| FACIT Fatigue | |||||||
| All | 3-month | 452 | −2.48 (0.66) | −8.77 (0.74) | −6.29 (0.99) | −5.32 (0.94) | 0.00000002 |
| All | 6-month | 466 | −1.97 (0.64) | −4.37 (0.61) | −2.40 (0.88) | −1.55 (0.83) | 0.06 |
| All | 12-month | 452 | −2.14 (0.70) | −4.01 (0.64) | −1.86 (0.95) | −1.01 (0.87) | 0.25 |
| All | 24-month | 382 | −1.49 (0.74) | −4.27 (0.82) | −2.77 (1.11) | −1.76 (1.03) | 0.09 |
| All | 36-month | 336 | −1.83 (0.81) | −3.67 (0.88) | −1.84 (1.19) | −0.90 (1.07) | 0.40 |
| Pre-Menopausal | 3-month | 152 | −3.87 (1.41) | −8.01 (1.13) | −4.14 (1.79) | −3.11 (1.64) | 0.06 |
| Pre-Menopausal | 6-month | 150 | −1.66 (1.19) | −3.26 (0.96) | −1.60 (1.51) | −0.82 (1.43) | 0.57 |
| Pre-Menopausal | 12-month | 149 | −1.32 (1.51) | −2.99 (1.14) | −1.67 (1.88) | −1.12 (1.64) | 0.50 |
| Pre-Menopausal | 24-month | 116 | −2.52 (1.60) | −2.45 (1.44) | 0.07 (2.16) | 1.02 (2.07) | 0.62 |
| Pre-Menopausal | 36-month | 102 | −2.11 (1.76) | −1.60 (1.71) | 0.51 (2.45) | 1.46 (2.12) | 0.49 |
| Post-Menopausal | 3-month | 300 | −1.87 (0.72) | −9.22 (0.96) | −7.35 (1.18) | −6.42 (1.14) | 0.00000004 |
| Post-Menopausal | 6-month | 316 | −2.10 (0.76) | −4.97 (0.77) | −2.87 (1.09) | −1.99 (1.02) | 0.05 |
| Post-Menopausal | 12-month | 303 | −2.52 (0.75) | −4.55 (0.76) | −2.03 (1.07) | −1.16 (1.02) | 0.26 |
| Post-Menopausal | 24-month | 266 | −1.09 (0.82) | −5.14 (1.00) | −4.05 (1.28) | −3.02 (1.17) | 0.01 |
| Post-Menopausal | 36-month | 234 | −1.71 (0.89) | −4.67 (1.00) | −2.95 (1.34) | −2.01 (1.22) | 0.10 |
| PROMIS Fatigue | |||||||
| All | 3-month | 446 | 1.70 (0.44) | 6.10 (0.50) | 4.39 (0.67) | 3.68 (0.63) | 0.00000001 |
| All | 6-month | 462 | 1.26 (0.44) | 3.51 (0.50) | 2.25 (0.66) | 1.52 (0.62) | 0.01 |
| All | 12-month | 442 | 1.45 (0.50) | 2.76 (0.53) | 1.31 (0.73) | 0.60 (0.67) | 0.37 |
| All | 24-month | 379 | 1.34 (0.58) | 3.35 (0.61) | 2.01 (0.85) | 1.11 (0.77) | 0.15 |
| All | 36-month | 330 | 1.42 (0.61) | 2.86 (0.64) | 1.44 (0.89) | 0.93 (0.80) | 0.25 |
| Pre-Menopausal | 3-month | 150 | 1.66 (0.85) | 7.34 (0.83) | 5.69 (1.19) | 4.18 (1.13) | 0.0003 |
| Pre-Menopausal | 6-month | 147 | 1.12 (0.74) | 3.50 (0.89) | 2.38 (1.16) | 0.85 (1.11) | 0.44 |
| Pre-Menopausal | 12-month | 144 | 0.41 (0.93) | 2.92 (0.95) | 2.51 (1.33) | 1.28 (1.18) | 0.28 |
| Pre-Menopausal | 24-month | 117 | 1.90 (1.12) | 2.68 (1.16) | 0.78 (1.61) | −0.74 (1.53) | 0.63 |
| Pre-Menopausal | 36-month | 98 | 0.73 (1.26) | 2.36 (1.06) | 1.63 (1.66) | 0.41 (1.52) | 0.79 |
| Post-Menopausal | 3-month | 296 | 1.72 (0.52) | 5.41 (0.62) | 3.69 (0.80) | 3.33 (0.76) | 0.00002 |
| Post-Menopausal | 6-month | 315 | 1.32 (0.55) | 3.52 (0.60) | 2.19 (0.81) | 1.83 (0.75) | 0.02 |
| Post-Menopausal | 12-month | 298 | 1.92 (0.59) | 2.67 (0.64) | 0.75 (0.87) | 0.25 (0.82) | 0.76 |
| Post-Menopausal | 24-month | 262 | 1.10 (0.69) | 3.68 (0.72) | 2.57 (1.00) | 1.97 (0.88) | 0.03 |
| Post-Menopausal | 36-month | 232 | 1.70 (0.70) | 3.09 (0.81) | 1.39 (1.06) | 1.21 (0.94) | 0.20 |
ACFB= average change from baseline; Raw Diff= Arm CT+E ACFB minus Arm E ACFB; LM Diff= estimated treatment difference (CT+E minus E) from linear regression of score at timepoint on treatment and baseline score; p.LM= p-value for treatment difference from linear model.
FACIT-ES menopause-by-treatment interaction: p=0.97, 0.41, 0.11, 0.02, 0.02, at 3, 6, 12, 24, 36 months
FACIT Fatigue menopause-by-treatment interactions: p=0.13, 0.49, 0.85, 0.06, 0.17, at 3, 6, 12, 24, 36 months
PROMIS Fatigue menopause by treatment interactions: p=0.42, 0.48, 0.34, 0.08, 0.60, at 3, 6, 12, 24, 36 months
Figure 2.
Endocrine and Chemoendocrine Arms, Change over 36 Months
Women receiving CT+E reported significantly greater increases in FACIT Fatigue Scale scores from baseline to 3 months, and approaching significantly greater fatigue at 6 months, compared to women on E; change scores were comparable between treatment arms at at later timepoints (Table 2). Change scores in a negative direction indicate more fatigue. Change scores by treatment arm converged over time because the CT+E arm reported decreased fatigue post-chemotherapy (Figure 2b). However, scores for the CT+E arm remained worse than baseline at all follow-up timepoints.
Women randomized to CT+E reported significantly greater increases in FACT-ES scores from baseline to 3 months compared to women randomized to E; change scores were not significantly different between treatment arms at later timepoints (Table 2). Change scores in a negative direction indicate more symptomatology. Both arms reported significantly more endocrine symptoms at all follow-up assessments compared to baseline (Figure 2c). For all three PROs, we observed similar result patterns using ITT analysis (supplemental materials).
Meaningful Change
We calculated the percentage of per-protocol participants whose symptoms meaningfully changed across assessments (Figure 3). Using a prior approach33, we conservatively used 0.5 SD as the threshold for meaningful change. The estimated SD of baseline PROMIS Fatigue was 8.2, so we defined ‘Better’ as decrease of >4.1 points, ‘Same’ as change within +/− 4.1, and ‘Worse’ as increase of >4.1. At 3 months, 59% of women receiving CT+E reported worse fatigue compared to 34% among E; at 6 months it was 47% (CT+E) versus 33% (E). The magnitude of difference between arms is lower later (e.g., 40% CT+E vs 36% E at 12 months). Nevertheless, a sizable proportion of women on both arms reported worsened fatigue at long-term follow-up (35–44%, 24 & 36 months). We observed similar results with FACIT-Fatigue. Using an estimated SD 9.4, for FACT-ES baseline scores (‘Better’:+>4.7 points; ‘Same’:+/− 4.7; ‘Worse’:- >4.1), women randomized to CT+E had worsened endocrine therapy-related symptoms at 3 and 6 months (50%, 52%) relative to E (39%, 44%, respectively).
Figure 3.
Endocrine and Chemoendocrine Arms: Meaningful Change
PROMIS Fatigue 7 Short Form: Better= decrease >4.1 points; Same= change +/− 4.1; Worse=increase >4.1.
FACIT Fatigue: Better=increase>5.1 points; Same= change +/− 5.1; Worse=decease > 5.1.
FACT-ES: Better= increase >4.7 points; Same= change +/− 4.7; Worse=decrease >4.7.
Differences by Menopausal Status
We examined symptom change scores by menopausal status. Fatigue trajectories by treatment appear to be different for pre- versus postmenopausal women (Figures 2a&b), with the effect larger and more persistent for postmenopausal women. Postmenopausal women appear to have had a larger influence on the overall treatment arm differences in fatigue changes from baseline to 3 months. Post-menopausal women in the CT+E arm reported significantly higher increases in fatigue compared to those in the E arm at 24 months. However, menopause-by-treatment interactions were non-significant at all timepoints for both fatigue measures (Table 2).
Endocrine symptom trajectories by treatment were also different for pre- versus postmenopausal women (Figure 2c), with the effect larger and increasing over time for post-menopausal women. Menopause-by-treatment interactions were significant at 24 and 36 months (Table 2). Post-menopausal women in the CT+E arm reported significantly higher increases in endocrine symptoms compared to those in E.
DISCUSSION
Women receiving treatment for early-stage breast cancer commonly report fatigue and endocrine symptoms. Chemotherapy-related fatigue is expected, given known mechanisms of action34, and often assumed to be reversible a sufficient time from completion. However, long-term data to demonstrate resolution has been lacking and complicated by receipt of radiation and endocrine therapy. Chemotherapy may also produce endocrine symptoms by inducing transient or persistent ovarian failure in premenopausal patients35. Similarly, tamoxifen and aromatase inhibitor (AI) side effects36 differ by menopausal status37.
TAILORx allowed examination of the unique contribution of chemotherapy to fatigue and endocrine symptoms, as well as symptom trajectories extending into post-treatment. Symptoms were greater at follow-up timepoints compared to baseline for both arms. Women on CT+E reported significantly greater increases in fatigue and endocrine symptoms during chemotherapy, compared to those on E. While endocrine therapy is assumed to be well-tolerated, a considerable proportion of women on both arms reported fatigue at long-term follow-up that exceeded a conservative threshold for meaningful worsening. At 12–36 months, increases in fatigue and endocrine symptoms were not significantly different between arms; the trajectories of women on CT+E converged with those on E. That there was some fatigue resolution in the chemoendocrine arm should be reassuring to women who may benefit from chemotherapy based on clinicopathologic features and RSs. For women who can safely skip chemotherapy, our findings on the significant, acute chemotherapy-related symptoms support the value of precision guided therapy sparing uneccesary toxicity.
Treatment arm fatigue trajectories appeared different for pre- versus postmenopausal women, with the effect larger and more persistent in the latter, although the differences did not reach statistical significance. The trajectories of endocrine symptoms by treatment also appeared different by menopausal status, with the effect larger and increasing over time for post-menopausal patients, and menopause-by-treatment interactions significant at later timepoints. Patients randomized to CT+E began endocrine therapy after completing chemotherapy. Therefore, endocrine symptoms would not develop in the CT+E arm until later timepoints.
These findings suggest earlier results demonstrating prior chemotherapy is associated with greater treatment side effect bother (which predicted higher risk of early AI discontinuation38) may be explained by more endocrine symptoms among women on CT+E. Yet, TAILORx demonstrated a significantly lower risk of early endocrine therapy discontinuation among women on CT+E39. While we speculated chronic symptom burden may diminish endocrine therapy tolerability40, results indicate further study is needed. Endocrine therapy adherence remains a complex challenge; interventions must be comprehensive41 and PROs have predictive value in identifying women at risk for early discontinuation38,42.
This study’s strengths include the randomized prospective design, long-term follow-up and well-validated measures. Limitations include missing data, including some attrition, which may have introduced bias (although we observed overall good retention). Sample characteristics were similar to the overall trial—supporting generalizability. The per-protocol analysis may introduce bias; however, our ITT analysis yielded similar results. Therapy regimens were selected using clinician judgment, which introduced variability. The majority of women randomized to chemotherapy received docetaxel-cyclophosphamide, so it is possible we underestimated symptom burden associated with other regimens. The amount of patients receiving particular endocrine treatments were not assessed at all time points. We were unable to evaluate the impact of tamoxifen versus AI treatment in our analyses of menopausal status subgoups. Lastly, we were unable to define the contribution of radiation or surgery.
Results demonstrate the: fatigue experienced acutely during chemotherapy and decreasing afterward, long-term endocrine symptom trajectories, and significant proportions of women with persistent symptoms. Our findings support the importance of providing long-term, symptom assessment and management. In quantifying the unique contributions of chemotherapy to fatigue and endocrine symptoms, study results add to the research identifying women with breast cancer unlikely to benefit substantially from chemotherapy relative to associated HRQoL impact. Findings illustrate the symptom burden that women with early stage HR+ breast cancer, and intermediate RSs, can be spared when electing to receive endocrine versus chemoendocrine therapy. They also provide valuable longitudinal data on the trajectories of common, distressing symptoms from the patient perspective.
Supplementary Material
Acknowledgements:
The authors acknowledge: Jeff Abrams, MD, Sheila Taube, PhD, Lori Minasian, MD, and Ann O’Mara, PhD, Una Hopkins, RN, DNP, ECOG-ACRIN Operations staff, the Cancer Trials Support Unit, and Mary Lou Smith.
This study was conducted by the ECOG-ACRIN Cancer Research Group (Peter J. O’Dwyer, MD & Mitchell D. Schnall, MD, PhD, Group Co-Chairs) and supported by the National Cancer Institute under award numbers U10CA180820, U10CA180794, UG1CA189828, UG1CA189859, UG1CA233160, UG1CA233320, U10CA180863, and UG1CA233277, and the Canadian Cancer Society grant #704970. Additional support was provided by the Breast Cancer Research Foundation, the Komen Foundation, the Breast Cancer Research Stamp issued by the U.S. Postal Service, and Genomic Health.
The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Dr. Gray reports funding from Genomic Health, Inc. for data collection and delivery. Dr. Wagner reports personal fees from Celgene, Inc. and Athenex, Inc. outside the submitted work, and grants from the NCI during the study period. Dr. Sparano reports holding a patent (WO 2009140304 A1) related to tests to predict chemotherapy responsiveness, and grants from NCI during the study period. Dr. Cella is an unpaid board member of the PROMIS Health Organization, reports NIH funding for PROMIS, and is President of FACIT.org.
Conflicts of Interest
Drs. Garcia, Gray, Sparano, Tevaarwerk, Carlos, Yanez, Gareen, Whelan, Sledge, Cella and Wagner report NCI grant funding during the study period. Dr. Wagner reports personal fees from Celgene, Inc. and Athenex, Inc. outside the submitted work. Dr. Sparano reports holding a patent (WO 2009140304 A1) related to tests to predict chemotherapy responsiveness. Dr. Cella is President of FACIT.org.
REFERENCES
- 1.Torre LA, Islami F, Siegel RL, Ward EM, and Jemal A, Global cancer in women: burden and trends. 2017, AACR. [DOI] [PubMed] [Google Scholar]
- 2.Plevritis SK, Munoz D, Kurian AW, et al. , Association of Screening and Treatment With Breast Cancer Mortality by Molecular Subtype in US Women, 2000–2012. JAMA, 2018. 319(2): p. 154–164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.ACS, Breast cancer facts & figures 2017–2018. 2017.
- 4.Davies C, Pan H, Godwin J, et al. , Long-term effects of continuing adjuvant tamoxifen to 10 years versus stopping at 5 years after diagnosis of oestrogen receptor-positive breast cancer: ATLAS, a randomised trial. Lancet, 2013. 381(9869): p. 805–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Goss PE, Ingle JN, Pritchard KI, et al. , Extending Aromatase-Inhibitor Adjuvant Therapy to 10 Years. New England Journal of Medicine, 2016. 375(3): p. 209–219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Cella D, Fallowfield L, Barker P, Cuzick J, Locker G, and Howell A, 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 Res Treat, 2006. 100(3): p. 273–84. [DOI] [PubMed] [Google Scholar]
- 7.Lash TL, Fox MP, Westrup JL, Fink AK, and Silliman RA, Adherence to tamoxifen over the five-year course. Breast Cancer Res Treat, 2006. 99(2): p. 215–20. [DOI] [PubMed] [Google Scholar]
- 8.Atkins L. and Fallowfield L, Intentional and non-intentional non-adherence to medication amongst breast cancer patients. Eur J Cancer, 2006. 42(14): p. 2271–6. [DOI] [PubMed] [Google Scholar]
- 9.Hershman DL, Shao T, Kushi LH, et al. , Early discontinuation and non-adherence to adjuvant hormonal therapy are associated with increased mortality in women with breast cancer. Breast Cancer Res Treat, 2011. 126(2): p. 529–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Shapiro CL and Recht A, Side effects of adjuvant treatment of breast cancer. N Engl J Med, 2001. 344(26): p. 1997–2008. [DOI] [PubMed] [Google Scholar]
- 11.Peto R, Davies C, Godwin J, et al. , Comparisons between different polychemotherapy regimens for early breast cancer: meta-analyses of long-term outcome among 100,000 women in 123 randomised trials. Lancet, 2012. 379(9814): p. 432–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Group EBCTC, Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet, 2005. 365(9472): p. 1687–717. [DOI] [PubMed] [Google Scholar]
- 13.Davies C, Godwin J, Gray R, et al. , Relevance of breast cancer hormone receptors and other factors to the efficacy of adjuvant tamoxifen: patient-level meta-analysis of randomised trials. Lancet, 2011. 378(9793): p. 771–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gray RG, Rea D, Handley K, et al. , aTTom: Long-term effects of continuing adjuvant tamoxifen to 10 years versus stopping at 5 years in 6,953 women with early breast cancer. Journal of Clinical Oncology, 2013. 31(18_suppl): p. 5–5.23169505 [Google Scholar]
- 15.Jin H, Tu D, Zhao N, Shepherd LE, and Goss PE, Longer-term outcomes of letrozole versus placebo after 5 years of tamoxifen in the NCIC CTG MA.17 trial: analyses adjusting for treatment crossover. J Clin Oncol, 2012. 30(7): p. 718–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sparano JA and Paik S, Development of the 21-gene assay and its application in clinical practice and clinical trials. J Clin Oncol, 2008. 26(5): p. 721–8. [DOI] [PubMed] [Google Scholar]
- 17.Sparano JA, Gray RJ, Makower DF, et al. , Prospective Validation of a 21-Gene Expression Assay in Breast Cancer. New England Journal of Medicine, 2015. 373(21): p. 2005–2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sparano JA, Gray RJ, Makower DF, et al. , Adjuvant Chemotherapy Guided by a 21-Gene Expression Assay in Breast Cancer. N Engl J Med, 2018. 379(2): p. 111–121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Sparano JA, Gray RJ, Makower DF, et al. , Clinical outcomes in early breast cancer with a high 21-gene recurrence score of 26 to 100 assigned to adjuvant chemotherapy plus endocrine therapy: A secondary analysis of the TAILORx randomized clinical trial. JAMA oncology, 2020. 6(3): p. 367–374. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Basch E, The Missing Voice of Patients in Drug-Safety Reporting. New England Journal of Medicine, 2010. 362(10): p. 865–869. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Patrick DL, Burke LB, Powers JH, et al. , Patient-reported outcomes to support medical product labeling claims: FDA perspective. Value Health, 2007. 10 Suppl 2: p. S125–37. [DOI] [PubMed] [Google Scholar]
- 22.Bower JE, Wiley J, Petersen L, Irwin MR, Cole SW, and Ganz PA, Fatigue after breast cancer treatment: Biobehavioral predictors of fatigue trajectories. Health Psychol, 2018. 37(11): p. 1025–1034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Garcia SF, Cella D, Clauser SB, et al. , Standardizing patient-reported outcomes assessment in cancer clinical trials: a patient-reported outcomes measurement information system initiative. J Clin Oncol, 2007. 25(32): p. 5106–12. [DOI] [PubMed] [Google Scholar]
- 24.Lai JS, Cella D, Choi S, et al. , How item banks and their application can influence measurement practice in rehabilitation medicine: a PROMIS fatigue item bank example. Arch Phys Med Rehabil, 2011. 92(10 Suppl): p. S20–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Yellen SB, Cella DF, Webster K, Blendowski C, and Kaplan E, Measuring fatigue and other anemia-related symptoms with the Functional Assessment of Cancer Therapy (FACT) measurement system. J Pain Symptom Manage, 1997. 13(2): p. 63–74. [DOI] [PubMed] [Google Scholar]
- 26.Fallowfield LJ, Leaity SK, Howell A, Benson S, and Cella D, Assessment of quality of life in women undergoing hormonal therapy for breast cancer: validation of an endocrine symptom subscale for the FACT-B. Breast Cancer Res Treat, 1999. 55(2): p. 189–99. [DOI] [PubMed] [Google Scholar]
- 27.Wagner L, Lai J, Cella D, Sweet J, and Forrestal S, Chemotherapy-related cognitive deficits: development of the FACT-Cog instrument. Ann Behav Med, 2004. 27(Suppl 10). [Google Scholar]
- 28.Jacobs SR, Jacobsen PB, Booth-Jones M, Wagner LI, and Anasetti C, Evaluation of the functional assessment of cancer therapy cognitive scale with hematopoietic stem cell transplant patients. J Pain Symptom Manage, 2007. 33(1): p. 13–23. [DOI] [PubMed] [Google Scholar]
- 29.Cella DF, Tulsky DS, Gray G, et al. , The Functional Assessment of Cancer Therapy scale: development and validation of the general measure. J Clin Oncol, 1993. 11(3): p. 570–9. [DOI] [PubMed] [Google Scholar]
- 30.Gotay CC and Pagano IS, Assessment of Survivor Concerns (ASC): a newly proposed brief questionnaire. Health Qual Life Outcomes, 2007. 5: p. 15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Wagner LI, Gray RJ, Sparano JA, et al. , Patient-Reported Cognitive Impairment Among Women With Early Breast Cancer Randomly Assigned to Endocrine Therapy Alone Versus Chemoendocrine Therapy: Results From TAILORx. Journal of Clinical Oncology, 2020. 38(17): p. 1875–1886. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Team, R.C., R: A language and environment for statistical computing. 2020, R Foundation for Statistical Computing: Vienna, Austria. [Google Scholar]
- 33.Sloan JA, Cella D, and Hays RD, Clinical significance of patient-reported questionnaire data: another step toward consensus. J Clin Epidemiol, 2005. 58(12): p. 1217–9. [DOI] [PubMed] [Google Scholar]
- 34.Baum M, Budzar AU, Cuzick J, et al. , Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomised trial. Lancet, 2002. 359(9324): p. 2131–9. [DOI] [PubMed] [Google Scholar]
- 35.Mauri D, Gazouli I, Zarkavelis G, et al. , Chemotherapy associated ovarian failure. Frontiers in Endocrinology, 2020. 11: p. 935. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Cuzick J, Sestak I, Forbes JF, et al. , Anastrozole for prevention of breast cancer in high-risk postmenopausal women (IBIS-II): an international, double-blind, randomised placebo-controlled trial. The Lancet, 2014. 383(9922): p. 1041–1048. [DOI] [PubMed] [Google Scholar]
- 37.Ferreira A, Di Meglio A, Pistilli B, et al. , Differential impact of endocrine therapy and chemotherapy on quality of life of breast cancer survivors: a prospective patient-reported outcomes analysis. Annals of Oncology, 2019. 30(11): p. 1784–1795. [DOI] [PubMed] [Google Scholar]
- 38.Wagner LI, Zhao F, Goss PE, et al. , Patient-reported predictors of early treatment discontinuation: treatment-related symptoms and health-related quality of life among postmenopausal women with primary breast cancer randomized to anastrozole or exemestane on NCIC Clinical Trials Group (CCTG) MA.27 (E1Z03). Breast Cancer Res Treat, 2018. 169(3): p. 537–548. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Yanez BG, Sparano RA, Carlos JA, et al. , Modifiable Predictors Associated with Early Discontinuation of Adjuvant Endocrine Therapy: Results from the TAILORx Trial. JAMA Oncology (In press).
- 40.Wagner LI, Patient-Reported Outcomes Bridge an Important Gap in Identifying Risk for Early Endocrine Therapy Discontinuation. Journal of the National Cancer Institute, 2021. [DOI] [PubMed]
- 41.Hershman DL, Unger JM, Hillyer GC, et al. , Randomized Trial of Text Messaging to Reduce Early Discontinuation of Adjuvant Aromatase Inhibitor Therapy in Women With Early-Stage Breast Cancer: SWOG S1105. J Clin Oncol, 2020. 38(19): p. 2122–2129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Hershman DL, Neugut AI, Moseley A, et al. , Patient Reported Outcomes and Long-Term Non-Adherence to Aromatase Inhibitors. J Natl Cancer Inst, 2021. [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.