Assessment of Quality of Life in MA.17: A Randomized, Placebo-Controlled Trial of Letrozole After 5 Years of Tamoxifen in Postmenopausal Wom
http://www.100md.com
《临床肿瘤学》
McMaster University, Hamilton
the National Cancer Institute of Canada Clinical Trials Group, Kingston
Toronto Sunnybrook Regional Cancer Centre, Toronto, ON, Canada
Harvard Medical School, Boston, MA
Mayo Clinic, Rochester, MN
Inova Fairfax Hospital, Falls Church, VA
John Wayne Cancer Institute, Santa Monica, CA
University of Vermont, Burlington, VT
ABSTRACT
PURPOSE: To evaluate the impact of letrozole compared with placebo after adjuvant tamoxifen on quality of life (QOL) in the MA.17 trial.
METHODS: Patients completed the Short Form 36-item Health Survey (SF-36) and the Menopause Specific Quality of Life Questionnaire (MENQOL) at baseline, 6 months, and annually. Mean change scores from baseline were compared between groups for summary measures and domains. A response analysis compared the proportion of patients who demonstrated an important change in QOL.
RESULTS: Of 5,187 randomly assigned women in the trial, 3,612 (69.9%) participated in the QOL substudy: 1,799 were allocated to placebo and 1,813 were allocated to letrozole. No differences were seen between groups in mean change scores from baseline for the SF-36 physical and mental component summary scores at 6, 12, 24, and 36 months. Small (< 0.2 standard deviations) but statistically significant differences in mean change scores from baseline were seen for the SF-36 domains of physical functioning (12 months), bodily pain (6 months) and vitality (6 and 12 months), and the MENQOL vasomotor (6, 12, and 24 months) and sexual domains (12 and 24 months). On the response analysis, a significant difference was seen between groups for the bodily pain domain (percentage of patients reporting a worsening of QOL, 47% placebo v 51% letrozole; P = .009) and the vasomotor domain (22% placebo v 29% letrozole; P = .001).
CONCLUSION: Letrozole did not have an adverse impact on overall QOL. Small effects were seen in some domains consistent with a minority of patients experiencing changes in QOL compatible with a reduction in estrogen synthesis.
INTRODUCTION
Adjuvant hormonal therapy results in substantial improvements in disease-free and overall survival for women with operable breast cancer. Five years of adjuvant tamoxifen has been the standard treatment in postmenopausal women. Recently, third-generation aromatase inhibitors have been tested in the adjuvant setting1-4 on the basis of positive results in metastatic disease.5-9 MA.17 is a placebo-controlled trial of the nonsteroidal aromatase inhibitor letrozole, 2.5 mg daily for 5 years, in postmenopausal women who were disease free after an initial 5 years of tamoxifen.4 After a median follow-up of 30 months, letrozole continues to demonstrate a significant improvement in disease-free survival, and an overall survival benefit in women with node-positive disease has been observed.10 The reduction in recurrence was seen for local and distant recurrences and for contralateral breast cancer.
More than 90% of the observed adverse effects in all randomly assigned women were grades 1 and 2. A statistically significant increase was seen in hot flashes, arthritis/arthralgia, and muscle pain in patients receiving letrozole compared with those receiving placebo. Vaginal bleeding was more common in patients receiving placebo. An increase in the percentage of patients reporting a new diagnosis of osteoporosis, which was not clinically verified, was also seen for those receiving letrozole.
An important consideration in clinical decision making regarding the use of aromatase inhibitors in the adjuvant setting is the potential impact on quality of life. In the recent Arimidex, Tamoxifen, Alone or in Combination (ATAC) trial, despite differences in reported adverse effects, no difference was noted in overall quality of life of patients receiving anastrozole versus tamoxifen.11 MA.17 is unique among current adjuvant trials in being able to evaluate an aromatase inhibitor compared with placebo in a group of well women free of recurrent cancer. Such women with early breast cancer, 5 years from diagnosis, were expected to have a relatively good quality of life. It was anticipated that the near-complete inhibition of estrogen synthesis provided by letrozole could have some detrimental adverse effects, possibly on bone and lipid metabolism and urogenital function, which might impair quality of life. Such effects have not been evaluated previously, given that a tool to ablate estrogen production and thereby examine its physiologic role has never been available. The MA.17 trial provided an opportunity to evaluate not only the clinical efficacy of letrozole against breast cancer, but also other effects of lowering estrogen.
PATIENTS AND METHODS
Design
Details of the methods of the randomized trial have been published previously4 and are outlined here briefly. From August 1998 to September 2002, postmenopausal women with primary breast cancer who had completed approximately 5 years of adjuvant tamoxifen were randomly assigned to receive daily oral letrozole 2.5 mg or placebo for 5 years. The multicenter trial involved the following cooperative groups: National Cancer Institute of Canada Clinical Trials Group (NCIC CTG), Southwest Oncology Group, Eastern Cooperative Oncology Group, Cancer and Leukemia Group B, North Central Cancer Treatment Group, European Organization for Research and Treatment of Cancer, and the International Breast Cancer Study Group. Women were stratified according to tumor hormone receptor status (positive or unknown), lymph node status (negative, positive, or unknown), and whether they received prior adjuvant chemotherapy. The ethics review board of each participating institution approved the main study protocol, including the quality-of-life substudy. All patients provided written informed consent. The primary end point for the main trial was disease-free survival. Secondary end points included overall survival, morbidity, and quality of life as presented here. The main study was terminated when a significant improvement in disease-free survival was observed at the first interim analysis.4
Patients
Women were eligible for the main trial if they were at least 50 years of age at the start of adjuvant tamoxifen therapy; if they were younger than 50 years but were postmenopausal or had undergone a bilateral oophorectomy before initiation of tamoxifen therapy, if they were younger than 50 years and premenopausal at the initiation of tamoxifen therapy but had become amenorrheic, on treatment with tamoxifen or if they had postmenopausal levels of luteinizing or follicle stimulating hormone. Other criteria included previous adjuvant tamoxifen therapy lasting 4.5 to 6 years, histologically confirmed diagnosis of primary hormone-positive breast cancer with no evidence of metastatic disease, discontinuation of tamoxifen less than 3 months before enrollment, Eastern Cooperative Oncology Group performance status of 0 to 2, and a life expectancy more than 5 years. Exclusion criteria included concomitant treatment with hormone replacement therapy or a selective estrogen receptor modulator. Intermittent treatment with vaginal estrogens, alternative medicines, and bisphosphonates was permitted. Eligibility for the quality-of-life substudy included willingness to complete quality-of-life questionnaires before random assignment and fluency in English or French. Participation in the quality-of-life substudy was part of the eligibility criteria for the main trial for all NCIC CTG centers unless patients were unable to complete questionnaires because of illiteracy, loss of sight, or other equivalent reasons. Participation was optional for patients enrolled through other cooperative groups.
Assessments
Quality of life was assessed using the Medical Outcomes Study 36-item Short Form Health Survey (SF-36)12,13 and the Menopause Specific Quality of Life Questionnaire (MENQOL).14 The instruments were administered at baseline before random assignment, at 6 months after random assignment, and then yearly. Both instruments have been shown in previous studies to be valid, reliable, and sensitive to change.12,15
The SF-36 is a multipurpose quality-of-life measure with 36 items. It contains eight subscales or domains, which are summarized into two global scores: the physical and mental component summary (PCS and MCS) scores.13 The results of the two summary scores provide a global indicator of patients' quality of life. Additional details on specific quality-of-life aspects may be ascertained using the individual eight subscales or domains, which include physical functioning (10 items), role-physical (four items), bodily pain (two items), general health (five items), vitality (four items), social functioning (two items), role-emotional (three items), and mental health (five items). This instrument has been used extensively in cancer and noncancer populations. It was considered an appropriate instrument for this study in view of the relatively healthy population studied with a similar potential for future cancer and noncancer outcomes. The SF-36 is scored from 0 to 100, with a higher number representing a better or favorable quality of life. The PCS and MCS are normalized so that the mean score for a representative sample of the US population is 50, with a standard deviation (SD) of 10.
The MENQOL is a validated tool that assesses the level of discomfort associated with symptoms of menopause.14 The instrument was chosen specifically to assess symptoms related to menopause or estrogen depletion that might be worsened by the use of aromatase inhibitors. The questionnaire consists of a 29-item symptom checklist and scale of level of discomfort. The MENQOL asks patients to indicate if they have a particular symptom (yes or no) and to evaluate the bother level of the symptom on a scale from 0 to 6. Symptoms include hot flashes, night sweats, feeling depressed, aching muscles or joints, difficulty sleeping, weight gain, dry skin, vaginal dryness, and sexual dysfunction. The 29 items yield four specific domains: vasomotor (three items), physical (seven items), psychosocial (16 items), and sexual (three items). Each item is scored from 1 to 8, with the final score for each domain representing a mean of the items in that domain. Higher scores on this scale indicate greater degree of symptom discomfort or less favorable quality of life.
Statistical Methods
The data were analyzed according to a prescribed framework developed by the Quality of Life Committee of the NCIC CTG.16 The approach is aimed at minimizing bias and emphasizing the clinical meaningfulness of the data. It involved four separate steps: a comparison of the completion rates for questionnaires between groups; a comparison of baseline scores between groups; a comparison of mean change of scores from baseline; and a comparison of the proportion of patients with improved, stable, or worsened scores between groups. The latter is called a response analysis and identifies the proportion of patients demonstrating a minimally important improvement or worsening of quality of life at any time during follow-up. The determination of how much change in any given instrument constitutes a minimally important difference in quality of life is the subject of intense study. Several approaches may be used based on the effect size, the SE of the measurement, or an external anchor such as patients' global assessment.17-19 Generally, a change of 5% to 10% of the scale breadth or 0.5 of the SD of the scale is perceivable to patients as meaningful.18,19 For this study we chose a conservative value of 5 points on the SF-36 and 0.5 points on the MENQOL as important changes in quality of life.17,19
In view of the large breadth of data available, several levels of analyses were performed. The main analysis was a between-group comparison of the proportion of patients who reported a change in quality of life at any time in any of the SF-36 summary measures or domains and the MENQOL domains. The response analysis was based on the proportion of patients observed to have an improved (important increase in score from baseline), stable (did not experience an important change in score from baseline), or worsened score (important decrease in score from baseline) over the course of the study. As additional analyses, we performed a logistic regression to evaluate baseline characteristics that predicted a minimally important worsening in quality of life at any time in follow-up for the PCS, the MCS, the SF-36 bodily pain domain, and the MENQOL vasomotor domain. Baseline characteristics evaluated included age (50 to 59 v < 49 years; 60 to 69 v < 49 years; and 70 v < 49 years), race (white v nonwhite), prior diagnosis of osteoporosis (yes v no), prior diagnosis of cardiovascular disease (yes v no), and baseline quality of life score as a continuous variable.
To gain a better understanding of what specific symptoms might be affected by letrozole, we also looked at individual items on the MENQOL questionnaire under the vasomotor and sexual domains and other items that might be affected by low circulating estrogen levels. These items included whether patients were bothered by hot flashes, night sweats, sweating, a change in sexual desire, vaginal dryness, avoiding intimacy, aching in muscles and joints, difficulty sleeping, depression, poor memory, or weight gain. The proportion of patients after random assignment who reported being very bothered by a symptom with a score of 6 to 8 on a scale ranging from 1 (no symptoms or not at all bothered) to 8 (extremely bothered) were compared between groups for each item.
Compliance rates with quality-of-life assessment were determined for all patients enrolled onto the quality-of-life subprotocol who completed a baseline questionnaire. These were based on the number of patients completing assessments at the designated time points divided by the total number of patients still on study and expected to complete an assessment. All patients who completed the baseline quality-of-life and a subsequent assessment were included in the response analysis. The Wilcoxon rank sum test was used to compare mean change of quality-of-life scores from baseline. A linear mixed model with time from random assignment as the only covariate was used as a repeated-measure analysis to test the significance for the effect of time on change in quality-of-life scores from baseline for each scale. The analysis was done separately for each treatment group because interactions between treatment and time were observed for some of the scales. The 2 test was used to compare the proportion of patients with improved, stable, or worsened scores. A corrected P value of less than .01 was used to determine significance in view of the multiple comparisons performed. The analysis is based on an updated median follow-up of 30 months.
RESULTS
Patient Demographics and Compliance Rates
Of the 5,187 women randomized in MA.17 trial, 3,612 (69.6%) participated in the quality-of-life subprotocol: 1,799 were allocated to placebo and 1,813 were allocated to receive letrozole. The patient groups were well balanced for baseline characteristics and were similar to those of patients in the main trial (Table 1). The majority were older than 60 years of age, white, and had been treated with breast-conserving surgery. Approximately 50% had node-negative disease and 43% had received adjuvant chemotherapy. A minority of patients had a previous diagnosis of osteoporosis or cardiovascular disease.
Compliance with the quality-of-life assessment was more than 90% for all time points and no difference was noted between study groups (Table 2). Compliance for different domains of quality of life was similar except for the MENQOL sexual domain, which ranged from 80% to 90%.
Baseline Scores
Mean baseline scores for the PCS and MCS were similar between groups (Table 3). The mean PCS was similar to or just below the norm of the US female population. The mean MCS was slightly above the norm for the US female population. Mean baseline scores for the SF-36 domains were similar between groups. Mean scores for the different domains were slightly higher than the norm for a female population of similar age.13 Mean baseline scores for the MENQOL domains were also similar between groups and relatively favorable (all mean scores were < 3).
Mean Change of Scores From Baseline
Mean change of scores from baseline for the PCS and MCS over time are shown in Figures 1 and 2. We observed a statistically significant decrease in mean change of scores from baseline to 6 months that continued on to a lesser degree to 36 months for both study groups for the PCS and the MCS (P < .0001 for each study group for both the PCS and the MCS, respectively; Table 4). No differences were observed between study groups for both the PCS and the MCS.
A similar decrease over time in mean change of scores from baseline was seen for both study groups for all of the domains of the SF-36 (Table 4). Small but statistically significant differences in mean of change of scores from baseline between study groups were seen for the physical functioning domain at 12 months (difference in mean change of scores [DIMCS] = 0.8; P = .006), the bodily pain domain at 6 months (DIMCS = 1.6; P = .007; Fig 3), and the vitality domain at 6 months (DIMCS = 1.8; P = .003) and 12 months (DIMCS = 1.8; P = .008), respectively. The observed differences were less than 2 points on the 100-point scale or less than 0.1 of an SD. No differences between groups were observed for the other SF-36 domains.
For the MENQOL questionnaire, the vasomotor domain improved over time for both study groups but this seemed to be delayed for patients treated with letrozole (P < .0001 for both study groups; Table 4; Fig 4). Statistically significant differences between groups in mean change of scores were seen at 6, 12, and 24 months (DIMCS = 0.3, P < .001; DIMCS = 0.4, P < .001; and DIMCS = 0.3, P < .001, respectively). For the sexual domain no changes were seen in the placebo group over time but there was a statistically significant worsening for the letrozole group (P < .0001; Table 4). Between-group comparisons were statistically different at 12 (DIMCS = 0.2; P < .001) and 24 months (DIMCS = 0.3; P = .004). The observed differences in these domains ranged from 0.2 to 0.4 points on the 7-point scale or less than 0.2 of an SD. No differences in mean change of scores from baseline between groups were observed for the other MENQOL domains.
Response Analysis
The results of the response analysis are listed in Table 5. For the PCS, the majority of patients experienced a stable (41% for placebo and 39% for the letrozole group) or worsened quality-of-life score (38% for placebo and 40% for the letrozole group) on at least one occasion. A minority of patients experienced an improved quality-of-life score (21% for each group) on at least one occasion. For the MCS, again, the majority of patients experienced a stable (43% for placebo and 41% for the letrozole group) or worsened quality-of-life score (33% for placebo and 36% for the letrozole group) on at least one occasion. A minority of patients experienced an improved quality-of-life score (24% for placebo and 23% for the letrozole group) on at least one occasion. No significant differences were detected between groups for the proportion of patients with improved, stable, or worsened scores for the PCS (P = .27) or the MCS (P = .085). With respect to individual domains of the SF-36, a significant difference between groups was seen in the response analysis for the bodily pain domain, for which 47% in the placebo group and 51% in the letrozole group reported a worsening in quality of life at some point during the study (P = .009). For the MENQOL instrument, only the vasomotor domain showed significant a difference in the response analysis between groups, with 22% of the placebo group and 29% of the letrozole group reporting worsening in quality of life related to vasomotor symptoms (P < .001).
Predictors for Quality of Life
Significant predictors for worsening in quality of life for SF-36 PCS included age (50 to 59 v < 49 years, odds ratio [OR] = 1.70, P = .0008; 60 to 69 v < 49 years, OR = 2.04, P = .0001; 70 v < 49 years, OR = 2.56, P < .0001) and baseline PCS score (OR = 1.03; P < .0001). Sole predictor for worsening of the SF-36 MCS was the baseline MCS score (OR = 1.05; P < .0001). With respect to the SF-36 bodily pain domain, predictors included age (50-59 v < 49 years, OR = 1.71, P = .0004; 60 to 69 v < 49 years, OR = 1.51, P = .007; 70 v < 49 years, OR = 1.72, P = .0006), baseline bodily pain score (OR = 1.02; P < .0001), and treatment (OR = 1.27; P = .002). Predictors for worsening in the MENQOL vasomotor domain included age ( 70 v < 49 years, OR = 0.54; P = .0003), baseline vasomotor score (OR = 0.86; P < .0001), and treatment (OR = 1.46; P < .0001).
Symptom Analysis
Table 6 demonstrates that symptoms were found to be bothersome in a minority of patients both in the placebo and letrozole groups. Significant differences between groups were seen only for hot flashes, which increased from 17% in the placebo group to 22% in the letrozole group (P = .0002), and sweating, which increased from 14% in the placebo group to 18% in the letrozole group (P = .003). Trends were seen for increases in night sweats, vaginal dryness, aching in muscles and joints, and difficulty sleeping in the letrozole group. No differences were observed between groups for the proportion of patients reporting a change in sexual desire, avoiding intimacy, poor memory, feeling depressed, or weight gain.
DISCUSSION
Treatment with letrozole after 5 years of tamoxifen has been shown to result in improved disease-free and distant disease-free survival.10 Letrozole as an effective aromatase inhibitor was also associated with a number of adverse effects attributed to low levels of circulating estrogen. An important consideration for clinical decision making regarding the use of this treatment is the potential effect on patients' quality of life. Our results demonstrate that letrozole compared with placebo did not have a detrimental effect on overall quality of life after 36 months of treatment. However, letrozole use did result in small differences in quality-of-life scores in a number of domains including physical function, bodily pain, vitality, vasomotor, and sexual. These differences are consistent with a small number of patients experiencing an important change in quality of life. This was confirmed in the response analysis, for which 4% and 7% of patients experienced worsening quality of life in the bodily pain and vasomotor domains, respectively. These observations are comparable with the reporting of toxicity from the trial, where it was observed that there was an increase in the proportion of patients reporting grade 1 and 2 adverse effects of arthritis/arthralgia, muscle pain, and hot flashes with the use of letrozole, suggesting that such a toxicity profile may result in a small proportion of patients experiencing a decrease in quality of life associated with such symptoms.
To date, three randomized trials have reported on quality of life and/or symptoms related to the use of aromatase inhibitors in the adjuvant setting. Fallowfield et al11 evaluated quality of life over a period of 24 months in postmenopausal women with early breast cancer after primary treatment who were treated with anastrozole (n = 335), tamoxifen (n = 347), or the combination (n = 339) in the ATAC trial. They observed that overall quality of life improved over time and there was no difference between groups. The investigators also observed a decrease in quality of life related to the endocrine subscale for all treatments but no difference was observed between treatment groups. On evaluation of specific symptoms, anastrozole seemed to result in a decrease in patients reporting cold sweats, vaginal discharge, irritation, and bleeding, and an increase in patients reporting vaginal dryness, pain, or discomfort on intercourse and loss of sexual interest.
Asmar et al20 reported on menopausal symptoms for the first 12 months in 997 postmenopausal women randomly assigned to exemestane or tamoxifen after primary surgical treatment. Exemestane was associated with a decrease in the proportion of patients reporting vaginal discharge and an increase in the proportion of patients reporting vaginal dryness and bone/muscle aches.
Fallowfield et al21 recently reported the results of the quality-of-life subprotocol of the Intergroup Exemestane Study in which postmenopausal women with early breast cancer who remained disease free after 2 to 3 years of tamoxifen were randomly assigned to additional tamoxifen or exemestane to complete a total of 5 years of adjuvant therapy. Of 4,742 women randomly assigned, 582 participated in the quality-of-life substudy. Patients were observed for 24 months. No difference was noted in overall quality of life or in the endocrine-specific subdomain. Endocrine symptoms, in particular vasomotor symptoms, improved over time in both groups. No significant difference was noted between groups for any endocrine symptoms aside from vaginal discharge, which was decreased in patients treated with exemestane.
Comparison of the MA.17 quality-of-life substudy with previous studies is difficult because of differences related to the timing of adjuvant treatment, the aromatase inhibitors evaluated, the comparison arms studied, and the instruments used to assess quality of life. It is of interest, however, that despite this variability, these trials are consistent in that third-generation aromatase inhibitors did not seem to have a negative effect on overall quality of life but did seem to affect specific vasomotor and/or gynecologic symptoms, the latter of which may be associated with problems with sexual function. An increase in bone or muscle aches was reported in two trials.
The lack of effect on overall quality of life by a hormonally active agent observed in this study has been observed in large trials of other hormonal agents in healthy women, including the National Surgical Adjuvant Breast and Bowel Project (NSABP) P1 trial, the Women's Health Initiative trial, and the Heart and Estrogen/Progesterone Replacement Study (HERA) trial. The NSABP prevention trial evaluated quality of life in pre- and postmenopausal women at high risk of developing breast cancer who were treated with either tamoxifen or placebo (N = 11,064).22 No differences were observed in overall quality of life. An increase in the proportion of patients reporting vasomotor and gynecologic symptoms and sexual dysfunction was reported in patients treated with tamoxifen. In the Women's Health Initiative trial, quality of life was evaluated in healthy postmenopausal women treated with estrogen plus progesterone (n = 8,506) or placebo (n = 8,102).23 This study demonstrated only minor effects on quality of life and hormone therapy was associated with less decline in specific domains related to physical functioning, bodily pain, and sleep disturbance at 1 year. At 3 years no differences were observed. In the HERA trial, an older population of women (mean age, 67 years) with coronary artery disease were randomized to hormone replacement therapy or placebo.24 Hormone therapy was associated with a slightly greater decline in physical function and energy. Patients treated with hormonal therapy had less severe depressive symptoms, especially those with menopausal symptoms at baseline.
Quality-of-life data are difficult for clinicians to interpret.17 The reasons for this are multifactorial and include a lack of familiarity with the instruments used and how they relate to clinical observations. This study used a unique response analysis to interpret differences in mean quality-of-life scores between treatment groups.16 This approach is particularly useful when small differences in mean scores are observed. We used a relatively modest change in score (5% to 7%) for the scales used to define a response. Our results support that the small differences in mean scores observed in some domains were consistent with a small difference in proportion of patients experiencing an important change in quality of life.
In addition to the main findings of the study, we observed that quality of life decreased slightly over time in both groups in the summary measures and in the majority of the domains. Whether this could be attributed to stopping tamoxifen, the effects of aging, or an elevated baseline associated with participating in the trial is unclear. This latter effect has been seen in other studies.22 Although we did observe differences in mean change scores between groups in some domains (eg, bodily pain and vasomotor), it is of interest that these differences seemed to be less evident with longer follow-up. Important secondary analyses demonstrated that older age was a predictor for worsening quality of life related to overall physical health and bodily pain. However, age older than 70 years was associated with less worsening in quality of life related to vasomotor symptoms; conversely, younger postmenopausal women were more likely to experience a worsening in quality of life attributed to this domain. The latter is relevant when aromatase inhibitors are administered to younger women because the absolute risk of impaired quality of life related to vasomotor symptoms may be greater.
In the analysis of specific symptoms, letrozole was associated with an increase in the proportion of patients reporting hot flashes and sweating. Importantly, no increase was observed in the proportion of patients reporting depression, weight gain, and poor memory. Although such analyses may have limited ability to detect subtle differences between groups, they do suggest that letrozole is unlikely in the short to midterm to cause major problems for patients with respect to such symptoms.
One of the strengths of this study is that it used a widely accepted generic instrument to evaluate overall quality of life, which permitted the assessment of specific domains, and an additional instrument to assess the endocrine related effects of aromatase inhibitors. This was a large study with limited missing data and follow-up extended beyond 36 months. In addition, it is unique in that it compared an aromatase inhibitor to a placebo. Given that all drug treatments are associated with adverse events, results in this trial showing no significant impact on overall quality of life confirm the favorable tolerability of agents such as letrozole. The limitations of this study are that it included primarily white North American women; generalizability to other races or nationalities is less clear. Follow-up is still relatively short in terms of planned duration of treatment and additional follow-up will be necessary to confirm the study findings.
How should the results of the MA.17 quality-of-life substudy be applied in practice? The main trial clearly demonstrates that extended adjuvant treatment with letrozole after tamoxifen results in improved disease-free survival irrespective of nodal status, and a preplanned subgroup analysis suggests an overall survival benefit for women with node-positive disease. The quality-of-life substudy demonstrates no major adverse affect on overall quality of life. Small differences in a number of domains associated with estrogen depletion (eg, vasomotor and bodily pain) were seen. Women and their physicians who are considering the use of letrozole after tamoxifen will need to weigh the potential benefits of reducing recurrence and contralateral breast cancer against any potential long-term risks such as osteoporosis. However, for the vast majority of women, short-term quality-of-life effects should not be a major factor in making such a treatment decision. A minority of women who are treated with letrozole may experience problems related to estrogen depletion that affect day-to-day function and do not improve over the course of treatment. For such women, other supportive therapy (or if the symptoms cannot be controlled, discontinuation of treatment) may be considered.
Authors' Disclosures of Potential Conflicts of Interest
Although all authors completed the disclosure declaration, the following authors or their immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
NOTES
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Baum M, Buzdar A, Cuzick J, et al: Tamoxifen Alone or in Combination Trialists' Group: 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 359:2131-2139, 2002
Boccardo F, Rubagotti A, Amoroso D, et al: Anastrozole appears to be superior to tamoxifen in women already receiving adjuvant tamoxifen treatment. Breast Cancer Res Treat 83:S6, 2003 (suppl 1)
Coombes RC, Hall E, Gibson LJ, et al: A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. N Engl J Med 350:1081-1092, 2004
Goss PE, Ingle JN, Martino S, et al: A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. N Engl J Med 349:1793-1802, 2003
Nabholtz JM, Buzdar A, Pollak M, et al: Anastrozole is superior to tamoxifen as first-line therapy for advanced breast cancer in postmenopausal women: Results of a North American multicenter randomized trial. J Clin Oncol 18 22:3758-3767, 2000
Bonneterre J, Thurlimann B, Robertson JFR, et al: Anastrozole versus tamoxifen as first-line therapy for advanced breast cancer in 668 postmenopausal women: Results of the tamoxifen or arimidex randomized group efficacy and tolerability study. J Clin Oncol 18:3748-3757, 2000
Dombernowsky P, Smith I, Falkson G, et al: Letrozole, a new oral aromatase inhibitor for advanced breast cancer: Double-blind randomized trial showing a dose effect and improved efficacy and tolerability compared with megestrol acetate. J Clin Oncol 16:453-461, 1998
Mouridsen H, Gershanovich M, Sun Y, et al: Superior efficacy of letrozole versus tamoxifen as first-line therapy for postmenopausal women with advanced breast cancer: Results of a phase III study of the International Letrozole Breast Cancer Group. J Clin Oncol 19:2596-2606, 2001
Dirix L, Piccart MJ, Lohrisch C, et al: Efficacy of and tolerance to exemestane versus tamoxifen in 1st line hormone therapy of postmenopausal metastatic breast cancer patients: A European Organization for the Research and Treatment of Cancer (EORTC Breast Group) phase II trial with Pharmacia and Upjohn. Proc Am Soc Clin Oncol 20:29a, 2001 (abstr 114)
Goss PE, Ingle JN, Martino S, et al: Letrozole following tamoxifen as extended adjuvant therapy in receptor positive breast cancer: NCIC CTG MA.17. J Natl Cancer Inst (in press)
Fallowfield L, Cella D, Cuzick J, et al: quality of life of postmenopausal women in the Arimidex, Tamoxifen, Alone or in Combination (ATAC) adjuvant breast cancer trial. J Clin Oncol 22:4261-4271, 2004
Ware JE, Snow KK Kosinksi M, et al: SF-36 Health Survey Manual and Interpretation Guide. Boston, MA, The Health Institute, 1993
Ware JE, Kosinski M, Keller SD. SF-36 Physical and Mental Summary Scales: A Users Manual (ed 3). Boston, MA, The Health Institute, New England Medical Centre, 1994
Hilditch JR, Lewis J, Peter A, et al: A menopause-specific quality of life questionnaire: Development and psychometric properties. Maturitas 24:161-175, 1996
Hilditch JR, Lewis J, Ross AH, et al: A comparison of the effects of oral conjugated equine estrogen and transdermal estradiol-17 beta combined with an oral progestin on quality of life in postmenopausal women. Maturitas 24:77-84, 1996
Osoba D, Bezjak A, Brundage M, et al: Analysis and interpretation of health-related quality-of-life data from clinical trials: Basic approach of the National Cancer Institute of Canada (NCIC) Clinical Trials Group (CTG). Eur J Cancer 41:280-287, 2005
Guyatt G, Osoba D, Wu A, et al: Methods to explain the clinical significance of health status measures. Mayo Clin Proc 77:371-383, 2002
Norman GR, Sloan JA, Wyrwich W: Interpretation of changes in health-related quality of life: The remarkable universality of half a standard deviation. Med Care 41:582-592, 2003
Wyrwich KW, Fihn SD, Tierney WM, et al: Clinically important changes in health-related quality of life for patients with chronic obstructive pulmonary disease: An Expert Consensus Panel Report. J Gen Intern Med 18:196-202, 2003
Asmar L, Cantrell J, Vukelja SJ, et al: A planned comparison of menopausal symptoms during the first year in 1,000 patients receiving either exemestane or tamoxifen in a double-blind adjuvant hormonal study. J Clin Oncol 22:6S 2004 (suppl; abstr 516)
Fallowfield LJ, Price MH, Hall E, et al: Intergroup exemestane study: Results of the quality of life sub-protocol. Breast Cancer Res Treat 88:S8, 2004 (suppl 1; abstr 4)
Day R, Ganz PA, Costantino JP, et al: 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. J Clin Oncol 17:2659-2669, 1999
Hays J, Ockene JK, Brunner RL, et al: Effects of estrogen plus progestin on health-related quality of life. N Engl J Med 348:1839-1854, 2003
Hlatky MA, Boothroyd D, Vittinghoff, et al: Quality-of-life and depressive symptoms in postmenopausal women after receiving hormone therapy: Results from the Heart and Estrogen/Progestin Replacement Study (HERS) Trial. JAMA 287:591-597, 2005(Timothy J. Whelan, Paul E)
the National Cancer Institute of Canada Clinical Trials Group, Kingston
Toronto Sunnybrook Regional Cancer Centre, Toronto, ON, Canada
Harvard Medical School, Boston, MA
Mayo Clinic, Rochester, MN
Inova Fairfax Hospital, Falls Church, VA
John Wayne Cancer Institute, Santa Monica, CA
University of Vermont, Burlington, VT
ABSTRACT
PURPOSE: To evaluate the impact of letrozole compared with placebo after adjuvant tamoxifen on quality of life (QOL) in the MA.17 trial.
METHODS: Patients completed the Short Form 36-item Health Survey (SF-36) and the Menopause Specific Quality of Life Questionnaire (MENQOL) at baseline, 6 months, and annually. Mean change scores from baseline were compared between groups for summary measures and domains. A response analysis compared the proportion of patients who demonstrated an important change in QOL.
RESULTS: Of 5,187 randomly assigned women in the trial, 3,612 (69.9%) participated in the QOL substudy: 1,799 were allocated to placebo and 1,813 were allocated to letrozole. No differences were seen between groups in mean change scores from baseline for the SF-36 physical and mental component summary scores at 6, 12, 24, and 36 months. Small (< 0.2 standard deviations) but statistically significant differences in mean change scores from baseline were seen for the SF-36 domains of physical functioning (12 months), bodily pain (6 months) and vitality (6 and 12 months), and the MENQOL vasomotor (6, 12, and 24 months) and sexual domains (12 and 24 months). On the response analysis, a significant difference was seen between groups for the bodily pain domain (percentage of patients reporting a worsening of QOL, 47% placebo v 51% letrozole; P = .009) and the vasomotor domain (22% placebo v 29% letrozole; P = .001).
CONCLUSION: Letrozole did not have an adverse impact on overall QOL. Small effects were seen in some domains consistent with a minority of patients experiencing changes in QOL compatible with a reduction in estrogen synthesis.
INTRODUCTION
Adjuvant hormonal therapy results in substantial improvements in disease-free and overall survival for women with operable breast cancer. Five years of adjuvant tamoxifen has been the standard treatment in postmenopausal women. Recently, third-generation aromatase inhibitors have been tested in the adjuvant setting1-4 on the basis of positive results in metastatic disease.5-9 MA.17 is a placebo-controlled trial of the nonsteroidal aromatase inhibitor letrozole, 2.5 mg daily for 5 years, in postmenopausal women who were disease free after an initial 5 years of tamoxifen.4 After a median follow-up of 30 months, letrozole continues to demonstrate a significant improvement in disease-free survival, and an overall survival benefit in women with node-positive disease has been observed.10 The reduction in recurrence was seen for local and distant recurrences and for contralateral breast cancer.
More than 90% of the observed adverse effects in all randomly assigned women were grades 1 and 2. A statistically significant increase was seen in hot flashes, arthritis/arthralgia, and muscle pain in patients receiving letrozole compared with those receiving placebo. Vaginal bleeding was more common in patients receiving placebo. An increase in the percentage of patients reporting a new diagnosis of osteoporosis, which was not clinically verified, was also seen for those receiving letrozole.
An important consideration in clinical decision making regarding the use of aromatase inhibitors in the adjuvant setting is the potential impact on quality of life. In the recent Arimidex, Tamoxifen, Alone or in Combination (ATAC) trial, despite differences in reported adverse effects, no difference was noted in overall quality of life of patients receiving anastrozole versus tamoxifen.11 MA.17 is unique among current adjuvant trials in being able to evaluate an aromatase inhibitor compared with placebo in a group of well women free of recurrent cancer. Such women with early breast cancer, 5 years from diagnosis, were expected to have a relatively good quality of life. It was anticipated that the near-complete inhibition of estrogen synthesis provided by letrozole could have some detrimental adverse effects, possibly on bone and lipid metabolism and urogenital function, which might impair quality of life. Such effects have not been evaluated previously, given that a tool to ablate estrogen production and thereby examine its physiologic role has never been available. The MA.17 trial provided an opportunity to evaluate not only the clinical efficacy of letrozole against breast cancer, but also other effects of lowering estrogen.
PATIENTS AND METHODS
Design
Details of the methods of the randomized trial have been published previously4 and are outlined here briefly. From August 1998 to September 2002, postmenopausal women with primary breast cancer who had completed approximately 5 years of adjuvant tamoxifen were randomly assigned to receive daily oral letrozole 2.5 mg or placebo for 5 years. The multicenter trial involved the following cooperative groups: National Cancer Institute of Canada Clinical Trials Group (NCIC CTG), Southwest Oncology Group, Eastern Cooperative Oncology Group, Cancer and Leukemia Group B, North Central Cancer Treatment Group, European Organization for Research and Treatment of Cancer, and the International Breast Cancer Study Group. Women were stratified according to tumor hormone receptor status (positive or unknown), lymph node status (negative, positive, or unknown), and whether they received prior adjuvant chemotherapy. The ethics review board of each participating institution approved the main study protocol, including the quality-of-life substudy. All patients provided written informed consent. The primary end point for the main trial was disease-free survival. Secondary end points included overall survival, morbidity, and quality of life as presented here. The main study was terminated when a significant improvement in disease-free survival was observed at the first interim analysis.4
Patients
Women were eligible for the main trial if they were at least 50 years of age at the start of adjuvant tamoxifen therapy; if they were younger than 50 years but were postmenopausal or had undergone a bilateral oophorectomy before initiation of tamoxifen therapy, if they were younger than 50 years and premenopausal at the initiation of tamoxifen therapy but had become amenorrheic, on treatment with tamoxifen or if they had postmenopausal levels of luteinizing or follicle stimulating hormone. Other criteria included previous adjuvant tamoxifen therapy lasting 4.5 to 6 years, histologically confirmed diagnosis of primary hormone-positive breast cancer with no evidence of metastatic disease, discontinuation of tamoxifen less than 3 months before enrollment, Eastern Cooperative Oncology Group performance status of 0 to 2, and a life expectancy more than 5 years. Exclusion criteria included concomitant treatment with hormone replacement therapy or a selective estrogen receptor modulator. Intermittent treatment with vaginal estrogens, alternative medicines, and bisphosphonates was permitted. Eligibility for the quality-of-life substudy included willingness to complete quality-of-life questionnaires before random assignment and fluency in English or French. Participation in the quality-of-life substudy was part of the eligibility criteria for the main trial for all NCIC CTG centers unless patients were unable to complete questionnaires because of illiteracy, loss of sight, or other equivalent reasons. Participation was optional for patients enrolled through other cooperative groups.
Assessments
Quality of life was assessed using the Medical Outcomes Study 36-item Short Form Health Survey (SF-36)12,13 and the Menopause Specific Quality of Life Questionnaire (MENQOL).14 The instruments were administered at baseline before random assignment, at 6 months after random assignment, and then yearly. Both instruments have been shown in previous studies to be valid, reliable, and sensitive to change.12,15
The SF-36 is a multipurpose quality-of-life measure with 36 items. It contains eight subscales or domains, which are summarized into two global scores: the physical and mental component summary (PCS and MCS) scores.13 The results of the two summary scores provide a global indicator of patients' quality of life. Additional details on specific quality-of-life aspects may be ascertained using the individual eight subscales or domains, which include physical functioning (10 items), role-physical (four items), bodily pain (two items), general health (five items), vitality (four items), social functioning (two items), role-emotional (three items), and mental health (five items). This instrument has been used extensively in cancer and noncancer populations. It was considered an appropriate instrument for this study in view of the relatively healthy population studied with a similar potential for future cancer and noncancer outcomes. The SF-36 is scored from 0 to 100, with a higher number representing a better or favorable quality of life. The PCS and MCS are normalized so that the mean score for a representative sample of the US population is 50, with a standard deviation (SD) of 10.
The MENQOL is a validated tool that assesses the level of discomfort associated with symptoms of menopause.14 The instrument was chosen specifically to assess symptoms related to menopause or estrogen depletion that might be worsened by the use of aromatase inhibitors. The questionnaire consists of a 29-item symptom checklist and scale of level of discomfort. The MENQOL asks patients to indicate if they have a particular symptom (yes or no) and to evaluate the bother level of the symptom on a scale from 0 to 6. Symptoms include hot flashes, night sweats, feeling depressed, aching muscles or joints, difficulty sleeping, weight gain, dry skin, vaginal dryness, and sexual dysfunction. The 29 items yield four specific domains: vasomotor (three items), physical (seven items), psychosocial (16 items), and sexual (three items). Each item is scored from 1 to 8, with the final score for each domain representing a mean of the items in that domain. Higher scores on this scale indicate greater degree of symptom discomfort or less favorable quality of life.
Statistical Methods
The data were analyzed according to a prescribed framework developed by the Quality of Life Committee of the NCIC CTG.16 The approach is aimed at minimizing bias and emphasizing the clinical meaningfulness of the data. It involved four separate steps: a comparison of the completion rates for questionnaires between groups; a comparison of baseline scores between groups; a comparison of mean change of scores from baseline; and a comparison of the proportion of patients with improved, stable, or worsened scores between groups. The latter is called a response analysis and identifies the proportion of patients demonstrating a minimally important improvement or worsening of quality of life at any time during follow-up. The determination of how much change in any given instrument constitutes a minimally important difference in quality of life is the subject of intense study. Several approaches may be used based on the effect size, the SE of the measurement, or an external anchor such as patients' global assessment.17-19 Generally, a change of 5% to 10% of the scale breadth or 0.5 of the SD of the scale is perceivable to patients as meaningful.18,19 For this study we chose a conservative value of 5 points on the SF-36 and 0.5 points on the MENQOL as important changes in quality of life.17,19
In view of the large breadth of data available, several levels of analyses were performed. The main analysis was a between-group comparison of the proportion of patients who reported a change in quality of life at any time in any of the SF-36 summary measures or domains and the MENQOL domains. The response analysis was based on the proportion of patients observed to have an improved (important increase in score from baseline), stable (did not experience an important change in score from baseline), or worsened score (important decrease in score from baseline) over the course of the study. As additional analyses, we performed a logistic regression to evaluate baseline characteristics that predicted a minimally important worsening in quality of life at any time in follow-up for the PCS, the MCS, the SF-36 bodily pain domain, and the MENQOL vasomotor domain. Baseline characteristics evaluated included age (50 to 59 v < 49 years; 60 to 69 v < 49 years; and 70 v < 49 years), race (white v nonwhite), prior diagnosis of osteoporosis (yes v no), prior diagnosis of cardiovascular disease (yes v no), and baseline quality of life score as a continuous variable.
To gain a better understanding of what specific symptoms might be affected by letrozole, we also looked at individual items on the MENQOL questionnaire under the vasomotor and sexual domains and other items that might be affected by low circulating estrogen levels. These items included whether patients were bothered by hot flashes, night sweats, sweating, a change in sexual desire, vaginal dryness, avoiding intimacy, aching in muscles and joints, difficulty sleeping, depression, poor memory, or weight gain. The proportion of patients after random assignment who reported being very bothered by a symptom with a score of 6 to 8 on a scale ranging from 1 (no symptoms or not at all bothered) to 8 (extremely bothered) were compared between groups for each item.
Compliance rates with quality-of-life assessment were determined for all patients enrolled onto the quality-of-life subprotocol who completed a baseline questionnaire. These were based on the number of patients completing assessments at the designated time points divided by the total number of patients still on study and expected to complete an assessment. All patients who completed the baseline quality-of-life and a subsequent assessment were included in the response analysis. The Wilcoxon rank sum test was used to compare mean change of quality-of-life scores from baseline. A linear mixed model with time from random assignment as the only covariate was used as a repeated-measure analysis to test the significance for the effect of time on change in quality-of-life scores from baseline for each scale. The analysis was done separately for each treatment group because interactions between treatment and time were observed for some of the scales. The 2 test was used to compare the proportion of patients with improved, stable, or worsened scores. A corrected P value of less than .01 was used to determine significance in view of the multiple comparisons performed. The analysis is based on an updated median follow-up of 30 months.
RESULTS
Patient Demographics and Compliance Rates
Of the 5,187 women randomized in MA.17 trial, 3,612 (69.6%) participated in the quality-of-life subprotocol: 1,799 were allocated to placebo and 1,813 were allocated to receive letrozole. The patient groups were well balanced for baseline characteristics and were similar to those of patients in the main trial (Table 1). The majority were older than 60 years of age, white, and had been treated with breast-conserving surgery. Approximately 50% had node-negative disease and 43% had received adjuvant chemotherapy. A minority of patients had a previous diagnosis of osteoporosis or cardiovascular disease.
Compliance with the quality-of-life assessment was more than 90% for all time points and no difference was noted between study groups (Table 2). Compliance for different domains of quality of life was similar except for the MENQOL sexual domain, which ranged from 80% to 90%.
Baseline Scores
Mean baseline scores for the PCS and MCS were similar between groups (Table 3). The mean PCS was similar to or just below the norm of the US female population. The mean MCS was slightly above the norm for the US female population. Mean baseline scores for the SF-36 domains were similar between groups. Mean scores for the different domains were slightly higher than the norm for a female population of similar age.13 Mean baseline scores for the MENQOL domains were also similar between groups and relatively favorable (all mean scores were < 3).
Mean Change of Scores From Baseline
Mean change of scores from baseline for the PCS and MCS over time are shown in Figures 1 and 2. We observed a statistically significant decrease in mean change of scores from baseline to 6 months that continued on to a lesser degree to 36 months for both study groups for the PCS and the MCS (P < .0001 for each study group for both the PCS and the MCS, respectively; Table 4). No differences were observed between study groups for both the PCS and the MCS.
A similar decrease over time in mean change of scores from baseline was seen for both study groups for all of the domains of the SF-36 (Table 4). Small but statistically significant differences in mean of change of scores from baseline between study groups were seen for the physical functioning domain at 12 months (difference in mean change of scores [DIMCS] = 0.8; P = .006), the bodily pain domain at 6 months (DIMCS = 1.6; P = .007; Fig 3), and the vitality domain at 6 months (DIMCS = 1.8; P = .003) and 12 months (DIMCS = 1.8; P = .008), respectively. The observed differences were less than 2 points on the 100-point scale or less than 0.1 of an SD. No differences between groups were observed for the other SF-36 domains.
For the MENQOL questionnaire, the vasomotor domain improved over time for both study groups but this seemed to be delayed for patients treated with letrozole (P < .0001 for both study groups; Table 4; Fig 4). Statistically significant differences between groups in mean change of scores were seen at 6, 12, and 24 months (DIMCS = 0.3, P < .001; DIMCS = 0.4, P < .001; and DIMCS = 0.3, P < .001, respectively). For the sexual domain no changes were seen in the placebo group over time but there was a statistically significant worsening for the letrozole group (P < .0001; Table 4). Between-group comparisons were statistically different at 12 (DIMCS = 0.2; P < .001) and 24 months (DIMCS = 0.3; P = .004). The observed differences in these domains ranged from 0.2 to 0.4 points on the 7-point scale or less than 0.2 of an SD. No differences in mean change of scores from baseline between groups were observed for the other MENQOL domains.
Response Analysis
The results of the response analysis are listed in Table 5. For the PCS, the majority of patients experienced a stable (41% for placebo and 39% for the letrozole group) or worsened quality-of-life score (38% for placebo and 40% for the letrozole group) on at least one occasion. A minority of patients experienced an improved quality-of-life score (21% for each group) on at least one occasion. For the MCS, again, the majority of patients experienced a stable (43% for placebo and 41% for the letrozole group) or worsened quality-of-life score (33% for placebo and 36% for the letrozole group) on at least one occasion. A minority of patients experienced an improved quality-of-life score (24% for placebo and 23% for the letrozole group) on at least one occasion. No significant differences were detected between groups for the proportion of patients with improved, stable, or worsened scores for the PCS (P = .27) or the MCS (P = .085). With respect to individual domains of the SF-36, a significant difference between groups was seen in the response analysis for the bodily pain domain, for which 47% in the placebo group and 51% in the letrozole group reported a worsening in quality of life at some point during the study (P = .009). For the MENQOL instrument, only the vasomotor domain showed significant a difference in the response analysis between groups, with 22% of the placebo group and 29% of the letrozole group reporting worsening in quality of life related to vasomotor symptoms (P < .001).
Predictors for Quality of Life
Significant predictors for worsening in quality of life for SF-36 PCS included age (50 to 59 v < 49 years, odds ratio [OR] = 1.70, P = .0008; 60 to 69 v < 49 years, OR = 2.04, P = .0001; 70 v < 49 years, OR = 2.56, P < .0001) and baseline PCS score (OR = 1.03; P < .0001). Sole predictor for worsening of the SF-36 MCS was the baseline MCS score (OR = 1.05; P < .0001). With respect to the SF-36 bodily pain domain, predictors included age (50-59 v < 49 years, OR = 1.71, P = .0004; 60 to 69 v < 49 years, OR = 1.51, P = .007; 70 v < 49 years, OR = 1.72, P = .0006), baseline bodily pain score (OR = 1.02; P < .0001), and treatment (OR = 1.27; P = .002). Predictors for worsening in the MENQOL vasomotor domain included age ( 70 v < 49 years, OR = 0.54; P = .0003), baseline vasomotor score (OR = 0.86; P < .0001), and treatment (OR = 1.46; P < .0001).
Symptom Analysis
Table 6 demonstrates that symptoms were found to be bothersome in a minority of patients both in the placebo and letrozole groups. Significant differences between groups were seen only for hot flashes, which increased from 17% in the placebo group to 22% in the letrozole group (P = .0002), and sweating, which increased from 14% in the placebo group to 18% in the letrozole group (P = .003). Trends were seen for increases in night sweats, vaginal dryness, aching in muscles and joints, and difficulty sleeping in the letrozole group. No differences were observed between groups for the proportion of patients reporting a change in sexual desire, avoiding intimacy, poor memory, feeling depressed, or weight gain.
DISCUSSION
Treatment with letrozole after 5 years of tamoxifen has been shown to result in improved disease-free and distant disease-free survival.10 Letrozole as an effective aromatase inhibitor was also associated with a number of adverse effects attributed to low levels of circulating estrogen. An important consideration for clinical decision making regarding the use of this treatment is the potential effect on patients' quality of life. Our results demonstrate that letrozole compared with placebo did not have a detrimental effect on overall quality of life after 36 months of treatment. However, letrozole use did result in small differences in quality-of-life scores in a number of domains including physical function, bodily pain, vitality, vasomotor, and sexual. These differences are consistent with a small number of patients experiencing an important change in quality of life. This was confirmed in the response analysis, for which 4% and 7% of patients experienced worsening quality of life in the bodily pain and vasomotor domains, respectively. These observations are comparable with the reporting of toxicity from the trial, where it was observed that there was an increase in the proportion of patients reporting grade 1 and 2 adverse effects of arthritis/arthralgia, muscle pain, and hot flashes with the use of letrozole, suggesting that such a toxicity profile may result in a small proportion of patients experiencing a decrease in quality of life associated with such symptoms.
To date, three randomized trials have reported on quality of life and/or symptoms related to the use of aromatase inhibitors in the adjuvant setting. Fallowfield et al11 evaluated quality of life over a period of 24 months in postmenopausal women with early breast cancer after primary treatment who were treated with anastrozole (n = 335), tamoxifen (n = 347), or the combination (n = 339) in the ATAC trial. They observed that overall quality of life improved over time and there was no difference between groups. The investigators also observed a decrease in quality of life related to the endocrine subscale for all treatments but no difference was observed between treatment groups. On evaluation of specific symptoms, anastrozole seemed to result in a decrease in patients reporting cold sweats, vaginal discharge, irritation, and bleeding, and an increase in patients reporting vaginal dryness, pain, or discomfort on intercourse and loss of sexual interest.
Asmar et al20 reported on menopausal symptoms for the first 12 months in 997 postmenopausal women randomly assigned to exemestane or tamoxifen after primary surgical treatment. Exemestane was associated with a decrease in the proportion of patients reporting vaginal discharge and an increase in the proportion of patients reporting vaginal dryness and bone/muscle aches.
Fallowfield et al21 recently reported the results of the quality-of-life subprotocol of the Intergroup Exemestane Study in which postmenopausal women with early breast cancer who remained disease free after 2 to 3 years of tamoxifen were randomly assigned to additional tamoxifen or exemestane to complete a total of 5 years of adjuvant therapy. Of 4,742 women randomly assigned, 582 participated in the quality-of-life substudy. Patients were observed for 24 months. No difference was noted in overall quality of life or in the endocrine-specific subdomain. Endocrine symptoms, in particular vasomotor symptoms, improved over time in both groups. No significant difference was noted between groups for any endocrine symptoms aside from vaginal discharge, which was decreased in patients treated with exemestane.
Comparison of the MA.17 quality-of-life substudy with previous studies is difficult because of differences related to the timing of adjuvant treatment, the aromatase inhibitors evaluated, the comparison arms studied, and the instruments used to assess quality of life. It is of interest, however, that despite this variability, these trials are consistent in that third-generation aromatase inhibitors did not seem to have a negative effect on overall quality of life but did seem to affect specific vasomotor and/or gynecologic symptoms, the latter of which may be associated with problems with sexual function. An increase in bone or muscle aches was reported in two trials.
The lack of effect on overall quality of life by a hormonally active agent observed in this study has been observed in large trials of other hormonal agents in healthy women, including the National Surgical Adjuvant Breast and Bowel Project (NSABP) P1 trial, the Women's Health Initiative trial, and the Heart and Estrogen/Progesterone Replacement Study (HERA) trial. The NSABP prevention trial evaluated quality of life in pre- and postmenopausal women at high risk of developing breast cancer who were treated with either tamoxifen or placebo (N = 11,064).22 No differences were observed in overall quality of life. An increase in the proportion of patients reporting vasomotor and gynecologic symptoms and sexual dysfunction was reported in patients treated with tamoxifen. In the Women's Health Initiative trial, quality of life was evaluated in healthy postmenopausal women treated with estrogen plus progesterone (n = 8,506) or placebo (n = 8,102).23 This study demonstrated only minor effects on quality of life and hormone therapy was associated with less decline in specific domains related to physical functioning, bodily pain, and sleep disturbance at 1 year. At 3 years no differences were observed. In the HERA trial, an older population of women (mean age, 67 years) with coronary artery disease were randomized to hormone replacement therapy or placebo.24 Hormone therapy was associated with a slightly greater decline in physical function and energy. Patients treated with hormonal therapy had less severe depressive symptoms, especially those with menopausal symptoms at baseline.
Quality-of-life data are difficult for clinicians to interpret.17 The reasons for this are multifactorial and include a lack of familiarity with the instruments used and how they relate to clinical observations. This study used a unique response analysis to interpret differences in mean quality-of-life scores between treatment groups.16 This approach is particularly useful when small differences in mean scores are observed. We used a relatively modest change in score (5% to 7%) for the scales used to define a response. Our results support that the small differences in mean scores observed in some domains were consistent with a small difference in proportion of patients experiencing an important change in quality of life.
In addition to the main findings of the study, we observed that quality of life decreased slightly over time in both groups in the summary measures and in the majority of the domains. Whether this could be attributed to stopping tamoxifen, the effects of aging, or an elevated baseline associated with participating in the trial is unclear. This latter effect has been seen in other studies.22 Although we did observe differences in mean change scores between groups in some domains (eg, bodily pain and vasomotor), it is of interest that these differences seemed to be less evident with longer follow-up. Important secondary analyses demonstrated that older age was a predictor for worsening quality of life related to overall physical health and bodily pain. However, age older than 70 years was associated with less worsening in quality of life related to vasomotor symptoms; conversely, younger postmenopausal women were more likely to experience a worsening in quality of life attributed to this domain. The latter is relevant when aromatase inhibitors are administered to younger women because the absolute risk of impaired quality of life related to vasomotor symptoms may be greater.
In the analysis of specific symptoms, letrozole was associated with an increase in the proportion of patients reporting hot flashes and sweating. Importantly, no increase was observed in the proportion of patients reporting depression, weight gain, and poor memory. Although such analyses may have limited ability to detect subtle differences between groups, they do suggest that letrozole is unlikely in the short to midterm to cause major problems for patients with respect to such symptoms.
One of the strengths of this study is that it used a widely accepted generic instrument to evaluate overall quality of life, which permitted the assessment of specific domains, and an additional instrument to assess the endocrine related effects of aromatase inhibitors. This was a large study with limited missing data and follow-up extended beyond 36 months. In addition, it is unique in that it compared an aromatase inhibitor to a placebo. Given that all drug treatments are associated with adverse events, results in this trial showing no significant impact on overall quality of life confirm the favorable tolerability of agents such as letrozole. The limitations of this study are that it included primarily white North American women; generalizability to other races or nationalities is less clear. Follow-up is still relatively short in terms of planned duration of treatment and additional follow-up will be necessary to confirm the study findings.
How should the results of the MA.17 quality-of-life substudy be applied in practice? The main trial clearly demonstrates that extended adjuvant treatment with letrozole after tamoxifen results in improved disease-free survival irrespective of nodal status, and a preplanned subgroup analysis suggests an overall survival benefit for women with node-positive disease. The quality-of-life substudy demonstrates no major adverse affect on overall quality of life. Small differences in a number of domains associated with estrogen depletion (eg, vasomotor and bodily pain) were seen. Women and their physicians who are considering the use of letrozole after tamoxifen will need to weigh the potential benefits of reducing recurrence and contralateral breast cancer against any potential long-term risks such as osteoporosis. However, for the vast majority of women, short-term quality-of-life effects should not be a major factor in making such a treatment decision. A minority of women who are treated with letrozole may experience problems related to estrogen depletion that affect day-to-day function and do not improve over the course of treatment. For such women, other supportive therapy (or if the symptoms cannot be controlled, discontinuation of treatment) may be considered.
Authors' Disclosures of Potential Conflicts of Interest
Although all authors completed the disclosure declaration, the following authors or their immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
NOTES
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Baum M, Buzdar A, Cuzick J, et al: Tamoxifen Alone or in Combination Trialists' Group: 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 359:2131-2139, 2002
Boccardo F, Rubagotti A, Amoroso D, et al: Anastrozole appears to be superior to tamoxifen in women already receiving adjuvant tamoxifen treatment. Breast Cancer Res Treat 83:S6, 2003 (suppl 1)
Coombes RC, Hall E, Gibson LJ, et al: A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. N Engl J Med 350:1081-1092, 2004
Goss PE, Ingle JN, Martino S, et al: A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. N Engl J Med 349:1793-1802, 2003
Nabholtz JM, Buzdar A, Pollak M, et al: Anastrozole is superior to tamoxifen as first-line therapy for advanced breast cancer in postmenopausal women: Results of a North American multicenter randomized trial. J Clin Oncol 18 22:3758-3767, 2000
Bonneterre J, Thurlimann B, Robertson JFR, et al: Anastrozole versus tamoxifen as first-line therapy for advanced breast cancer in 668 postmenopausal women: Results of the tamoxifen or arimidex randomized group efficacy and tolerability study. J Clin Oncol 18:3748-3757, 2000
Dombernowsky P, Smith I, Falkson G, et al: Letrozole, a new oral aromatase inhibitor for advanced breast cancer: Double-blind randomized trial showing a dose effect and improved efficacy and tolerability compared with megestrol acetate. J Clin Oncol 16:453-461, 1998
Mouridsen H, Gershanovich M, Sun Y, et al: Superior efficacy of letrozole versus tamoxifen as first-line therapy for postmenopausal women with advanced breast cancer: Results of a phase III study of the International Letrozole Breast Cancer Group. J Clin Oncol 19:2596-2606, 2001
Dirix L, Piccart MJ, Lohrisch C, et al: Efficacy of and tolerance to exemestane versus tamoxifen in 1st line hormone therapy of postmenopausal metastatic breast cancer patients: A European Organization for the Research and Treatment of Cancer (EORTC Breast Group) phase II trial with Pharmacia and Upjohn. Proc Am Soc Clin Oncol 20:29a, 2001 (abstr 114)
Goss PE, Ingle JN, Martino S, et al: Letrozole following tamoxifen as extended adjuvant therapy in receptor positive breast cancer: NCIC CTG MA.17. J Natl Cancer Inst (in press)
Fallowfield L, Cella D, Cuzick J, et al: quality of life of postmenopausal women in the Arimidex, Tamoxifen, Alone or in Combination (ATAC) adjuvant breast cancer trial. J Clin Oncol 22:4261-4271, 2004
Ware JE, Snow KK Kosinksi M, et al: SF-36 Health Survey Manual and Interpretation Guide. Boston, MA, The Health Institute, 1993
Ware JE, Kosinski M, Keller SD. SF-36 Physical and Mental Summary Scales: A Users Manual (ed 3). Boston, MA, The Health Institute, New England Medical Centre, 1994
Hilditch JR, Lewis J, Peter A, et al: A menopause-specific quality of life questionnaire: Development and psychometric properties. Maturitas 24:161-175, 1996
Hilditch JR, Lewis J, Ross AH, et al: A comparison of the effects of oral conjugated equine estrogen and transdermal estradiol-17 beta combined with an oral progestin on quality of life in postmenopausal women. Maturitas 24:77-84, 1996
Osoba D, Bezjak A, Brundage M, et al: Analysis and interpretation of health-related quality-of-life data from clinical trials: Basic approach of the National Cancer Institute of Canada (NCIC) Clinical Trials Group (CTG). Eur J Cancer 41:280-287, 2005
Guyatt G, Osoba D, Wu A, et al: Methods to explain the clinical significance of health status measures. Mayo Clin Proc 77:371-383, 2002
Norman GR, Sloan JA, Wyrwich W: Interpretation of changes in health-related quality of life: The remarkable universality of half a standard deviation. Med Care 41:582-592, 2003
Wyrwich KW, Fihn SD, Tierney WM, et al: Clinically important changes in health-related quality of life for patients with chronic obstructive pulmonary disease: An Expert Consensus Panel Report. J Gen Intern Med 18:196-202, 2003
Asmar L, Cantrell J, Vukelja SJ, et al: A planned comparison of menopausal symptoms during the first year in 1,000 patients receiving either exemestane or tamoxifen in a double-blind adjuvant hormonal study. J Clin Oncol 22:6S 2004 (suppl; abstr 516)
Fallowfield LJ, Price MH, Hall E, et al: Intergroup exemestane study: Results of the quality of life sub-protocol. Breast Cancer Res Treat 88:S8, 2004 (suppl 1; abstr 4)
Day R, Ganz PA, Costantino JP, et al: 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. J Clin Oncol 17:2659-2669, 1999
Hays J, Ockene JK, Brunner RL, et al: Effects of estrogen plus progestin on health-related quality of life. N Engl J Med 348:1839-1854, 2003
Hlatky MA, Boothroyd D, Vittinghoff, et al: Quality-of-life and depressive symptoms in postmenopausal women after receiving hormone therapy: Results from the Heart and Estrogen/Progestin Replacement Study (HERS) Trial. JAMA 287:591-597, 2005(Timothy J. Whelan, Paul E)