Menopausal symptoms in women undergoing chemotherapy-induced and natural menopause: a prospective controlled study

Menopausal symptoms in women undergoing chemotherapy-induced and natural menopause: a prospective controlled study

original article Annals of Oncology 21: 983–987, 2010 doi:10.1093/annonc/mdp394 Published online 14 October 2009 Menopausal symptoms in women underg...

73KB Sizes 0 Downloads 4 Views

original article

Annals of Oncology 21: 983–987, 2010 doi:10.1093/annonc/mdp394 Published online 14 October 2009

Menopausal symptoms in women undergoing chemotherapy-induced and natural menopause: a prospective controlled study H. G. Mar Fan1,2, N. Houe´de´-Tchen1,3, I. Chemerynsky1, Q.-L. Yi1,4, W. Xu1, B. Harvey5  & I. F. Tannock1* 1

Background: Women with breast cancer frequently undergo menopause following adjuvant chemotherapy. Here, we investigated whether they have more severe symptoms than women undergoing natural menopause.

Patients and methods: Forty-one women who had undergone menopause as a result of chemotherapy and 57 healthy women who had undergone recent natural menopause were evaluated on two occasions 1 year apart. The primary end point was the summed score of the self-report Functional Assessment of Cancer Therapy, endocrine symptoms (FACT-ES) scale. Quality of life was evaluated by the FACT-G questionnaire and fatigue by the FACT-F subscale. Results: There was a strong trend for patients to report worse FACT-ES scores than controls at the first (P = 0.05) and second (P = 0.04) time points. More patients had moderate/severe hot flashes than controls undergoing natural menopause (51% versus 19%, P = 0.003). Patients reported worse fatigue than controls at the first assessment (P = 0.04), with no difference at the second. Menopausal symptoms were associated with fatigue for both groups. There was no difference between patients and controls in the quality-of-life scale, although assessment of patients is likely subject to adaptation and response-shift bias. Conclusions: Women undergoing chemotherapy-induced menopause may experience worse symptoms than women undergoing natural menopause. Key words: breast cancer, chemotherapy, fatigue, menopause, quality of life

introduction Premenopausal women who receive adjuvant chemotherapy for early breast cancer may undergo premature menopause. Chemotherapy has been reported to cause early menopause in about 40% of 40-year-old and close to 100% of 50-year-old women [1]. The implications are wide ranging and include loss of fertility and accelerated bone loss. It was also our clinical impression that women with chemotherapy-induced menopause experience more frequent and severe symptoms, such as hot flashes, than women undergoing natural menopause. If this is true, it will be important not only to advise young women about to receive chemotherapy of the probability of early menopause but also that their symptoms may be more severe than those associated with a later natural menopause. *Correspondence to: Dr I. F. Tannock, Division of Medical Oncology and Hematology, Princess Margaret Hospital and University of Toronto, 610 University Avenue, Suite 5-218, Toronto, Ontario, Canada M5G 2M9. Tel: +1-416-946-2245; Fax: +1-416-946-6546; E-mail: [email protected]  

Formerly with the Ontario Breast Screening Program, Toronto, Ontario, Canada.

We had conducted previously a longitudinal study of menopausal symptoms, fatigue and cognitive dysfunction in women undergoing adjuvant chemotherapy for early breast cancer compared with healthy aged-matched controls [2, 3]. Many of the women with breast cancer experienced chemotherapy-induced menopause during this study, and their menopausal symptoms were documented and quantified [2, 3]; however, in that study we were unable to compare the effects of chemotherapy-induced and natural menopause as the majority of age-matched controls remained premenopausal. The study also included women who had undergone natural menopause before receiving chemotherapy. We therefore designed a companion study in which an additional cohort of healthy women undergoing natural menopause was recruited to allow comparison with women who had undergone menopause as a result of chemotherapy. This enabled comparison of the severity of menopausal symptoms and also allowed insight into the degree to which menopause is associated with fatigue, which is experienced by many women undergoing adjuvant chemotherapy.

ª The Author 2009. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected]

original article

Received 30 June 2008; revised 16 April 2009 & 17 June 2009; accepted 8 July 2009

Downloaded from http://annonc.oxfordjournals.org/ at Northern Arizona University on June 9, 2015

Division of Medical Oncology and Hematology, Princess Margaret Hospital and University of Toronto, Toronto, Ontario, Canada; 2Department of Medical Oncology, Princess Alexandra Hospital, Brisbane, Queensland, Australia; 3Department of Medical Oncology, Institut Bergonie´, Bordeaux, France; 4Department of Biostatistics, Canadian Blood Services, Ottawa, Ontario, Canada and 5Department of Family and Community Medicine, University of Toronto, Toronto, Ontario, Canada

original article patients and methods study population

assessments At each assessment, patients and controls were asked to complete a questionnaire about demographic and medical information in addition to menstrual history. They also completed a series of Functional Assessment of Cancer Therapy (FACT) self-assessment questionnaires for menopausal symptoms (FACT-ES) [4], fatigue (FACT-F) [5] and general quality of life (FACT-G) [6]. The FACT-ES is an 18-item subscale evaluating commonly reported menopausal symptoms, including vasomotor symptoms, vaginal symptoms and sexual dysfunction (see Table 3 for items addressed). It was validated in 268 women receiving endocrine therapy for breast cancer [4]. Patients were assessed toward the end of their chemotherapy (assessment 0) and at 1- and 2-year follow-up intervals (assessments 1 and 2). Controls were assessed on two occasions 1 year apart (assessments 1 and 2). Comparisons were made (i) between the patients at their 1-year follow-up assessment (generally about 9–12 months after the end of their chemotherapy) and the first assessment of controls (assessment 1), and (ii) the assessment of patients and controls 1 year (63 months) later (assessment 2). Data describing menopausal symptoms during or shortly after chemotherapy for patients are also presented (assessment 0); these data are presented as they were collected in our previously reported study and provide a baseline for the patients. As it was not possible to identify a cohort of control women in the very narrow time frame of peri-/early menopause, no comparison data are presented for this assessment.

statistical analysis The primary end point of the study was the summed FACT-ES score, and the primary comparison was between the distribution of total FACT-ES scores of patients and controls at assessment 1. We also evaluated differences between patients and controls for the distribution of total FACT-F and FACT-G scores. Because data were not normally distributed,

984 | Mar Fan et al.

the Wilcoxon rank sum test was used to compare the scores of the two groups. Differences in the primary end point were regarded as significant for a two-sided P value of £0.05. In view of the multiple comparisons, evidence for differences between groups for other end points has been regarded as significant for two-sided P value of £0.01. Individual menopausal symptoms and components of fatigue, defined by the FACT-ES and FACT-F, respectively, are presented using descriptive statistics. The following factors were entered into stepwise regression analyses for their influence on FACT-ES scores for patients: receipt of adjuvant hormonal therapy (or not) after chemotherapy, treatment of hot flashes, time since last period and FACT-F score. The following factors were entered into stepwise regression analyses for their influence on FACT-ES scores for controls: receipt of HRT, use of selective serotonin reuptake inhibitors (SSRI), time since last period and FACT-F score. Exploratory analyses were carried out for the influence of the selected variables on the FACT-F and FACT-G scores. The paired Wilcoxon rank test was applied to compare the change of the scores evaluated at the two time points. We had estimated that 50 patients from our previous study with chemotherapy-induced menopause would be eligible for the current study. For our control group, we aimed originally to recruit 100 women to give a case–control ratio of 1:2, but we subsequently revised this to include a minimum of 50 control women in view of difficulties in identifying and recruiting them.

results patients and controls Of 104 women with breast cancer assessed in our previous study, 66 were premenopausal at baseline [2]. Of initially premenopausal women with breast cancer, chemotherapyinduced menopause occurred in 31% of those aged <40 years, 53% of those aged 40–50 years, and 85% of those >50 years. In total, 41 of these women were postmenopausal at the time of assessment 1 (i.e. 9–12 months after the end of their chemotherapy) and were included in the current analysis. Thirty-five of these women underwent assessment 2. A total of 57 controls were recruited and assessed. Of these, 2 were eligible controls from our previous study and 55 were recruited from the OBSP database. To recruit these controls, a total of 2913 women from the OBSP database were screened. The most frequent reason for ineligibility was that cessation of periods had occurred outside the range of 6 months to 2 years. Of the 57 participating controls, 50 agreed to be reassessed 1 year later. As expected, patients and controls differed in age, with patients being significantly younger than controls (Table 1). Patients and controls were well matched for time from last period until the time of their first comparison of symptoms. The current use of drugs, which might impact on menopausal symptoms, including adjuvant hormonal therapy for breast cancer, HRT and other treatments of hot flashes, is also indicated in Table 1. Use of these agents did not change significantly between assessments. menopausal symptoms There was an effect of borderline statistical significance for patients to experience worse menopausal symptoms than controls at their first (P = 0.05) and second (P = 0.04) comparisons, and a substantially higher proportion of patients

Volume 21 | No. 5 | May 2010

Downloaded from http://annonc.oxfordjournals.org/ at Northern Arizona University on June 9, 2015

The group of women with chemotherapy-induced menopause was drawn from our previously reported study [2, 3]: they were included if their last menstrual period was <6 months before their first assessment (undertaken toward the end of their chemotherapy) and >6 months before their 1 year (6 3 months) follow-up assessment (i.e. the primary time point for assessment of menopausal symptoms in the present study). Patients who had recurrent breast cancer at one of their follow-up visits were excluded. The control population was of healthy women who had undergone recent menopause as defined by cessation of menses at least 6 months and not more than 2 years before enrolment. The initial assessment was scheduled within 2 months of recruitment. They were aged £60 years and were fluent in English. Exclusion criteria included a history of psychiatric disease, drug or alcohol abuse, hysterectomy and/or oophorectomy, major chronic illness, previous cancer and use of medication that might lead to cognitive problems. Controls who were receiving hormone replacement therapy (HRT) were included, as were patients who were receiving treatment for hot flashes. Eligible controls from our previous study were included. In addition, controls were recruited from the database of the Ontario Breast Screening Program (OBSP). We were provided with contact details of women aged 50–60 years, living in or close to Toronto, who had indicated a willingness to be contacted for breast-related studies. A cover letter, study information sheet and consent form were mailed to these women and followed with a telephone call. The protocol for this study was approved by the review boards of the participating institutions, and all participants gave written informed consent.

Annals of Oncology

original article

Annals of Oncology

Table 1. Demographic characteristics, and receipt of drugs that might influence menopausal symptoms, for patients and controls at their first comparative assessment Patient (n = 41)

Control (n = 57)

P value

48 (45–52) 15.6 (13.3–17.8)

54 (53–55) 15.9 (11.0–21.7)

<0.0001 0.75

27 5 5 4

NA

2 (7) 27 (93) 28 2 12

14 (25) 42 (75) NA 12 4

0.08

0.02 0.003

a

Twenty-three patients were receiving tamoxifen, 4 an aromatase inhibitor and 1 goserelin. Nine patients and 4 controls were taking a selective serotonin reuptake inhibitor; 2 patients were receiving clonidine and 1 evening primrose oil. C, cyclophosphomide; E, epirubicin; F, 5-fluorouracil; A, Adriamycin (doxorubicin); M, methotrexate; NA, not applicable.

b

Table 2. Menopausal symptoms for patients and controls

Assessment 0 Median FACT-ES score (interquartile range) Assessment 1 Median FACT-ES score (interquartile range) % with moderate/severe hot flashes Assessment 2 Median FACT-ES score (interquartile range) % with moderate/severe hot flashes

Table 3. Comparison of individual symptoms assessed by the FACT-ES scale for patients and controls at assessments 1 and 2

Patient

Control

P value

58 (50.5–61.5)

NA

58 (54–62)

62 (56–65)

0.05

51.4

19.3

0.003

61 (51–62)

62 (57–67)

0.04

27.3

10.4

0.002

Median FACT-ES scores (range 0–72) are indicated. Higher scores represent fewer symptoms. P values for comparison of the FACT-ES scores and for distribution of hot flashes between patients and controls are determined using the Wilcoxon log-rank test. FACT-ES, Functional Assessment of Cancer Therapy, endocrine symptoms subscale.

experienced moderate or severe hot flashes (Table 2). Differences in individual symptoms explored by the FACT-ES scale are shown in Table 3. In addition to hot flashes, patients tended to have more vaginal discharge and to have lost interest in sex as compared with controls.

fatigue and quality of life Despite improving from the assessment carried out near the end of chemotherapy, there was a trend for women with breast cancer to report more fatigue than controls at assessment 1 (Table 4). This difference had resolved by the time of assessment 2.

Volume 21 | No. 5 | May 2010

Hot flashes Cold sweats Night sweats Vaginal discharge Vaginal itching/irritation Vaginal bleeding/spotting Vaginal dryness Pain on intercourse Lost interest in sex Weight gain Lightheaded/dizzy Vomiting Diarrhea Headaches Feeling bloated Breast tenderness Mood swings Irritable Total FACT-ES score

Assessment 1

Assessment 2

0.003 NS NS 0.005 NS NS NS NS 0.03 NS 0.03 NS NS 0.09 0.05 0.06 0.05 NS 0.05

0.002 NS NS 0.02 NS NS NS NS 0.003 0.005 NS NS NS 0.07 NS 0.01 NS 0.07 0.04

P values (determined by the Wilcoxon log-rank test) are indicated if P <0.10 to show trends (and all trends are to worse symptoms for the patients). Other comparisons were not significant (NS). FACT-ES, Functional Assessment of Cancer Therapy, endocrine symptoms subscale.

Overall quality of life was not significantly different for patients and controls at the time of either comparison (Table 4).

multivariate analysis and time trends For both patients and controls, an association was found between menopausal symptoms (FACT-ES score) and fatigue

doi:10.1093/annonc/mdp394 | 985

Downloaded from http://annonc.oxfordjournals.org/ at Northern Arizona University on June 9, 2015

Age in years, median (interquartile range) Months since last period at initial assessment, median (interquartile range) Chemotherapy drugs received CEF CMF AC Others Educational level, n (%) Secondary Postsecondary Adjuvant hormonal treatmenta Hormone replacement therapy Nonhormonal treatment of hot flashesb

original article

Annals of Oncology

Table 4. Fatigue and quality of life reported by patients and controls

Assessment 0, median (IQR) Assessment 1, median (IQR) Assessment 2, median (IQR)

Fatigue (FACT-F) Patient

Control

P value

Quality of life (FACT-G) Patient Control

P value

31 (20–38) 42 (38–47) 45 (39–48)

NA 46 (41–50) 45 (41–50)

0.04 0.22

75 (66–85) 88 (79–96) 87 (81–97)

0.49 1.00

NA 91 (80–97) 90 (76–96)

Median FACT-F (range 0–52) and FACT-G (range 0–108) scores are indicated. Higher scores indicate less fatigue or better quality of life. Baseline assessment for patients was carried out during or shortly after chemotherapy. FACT, Functional Assessment of Cancer Therapy (F = fatigue subscale; G = general); IQR, interquartile range.

discussion Chemotherapy-induced menopause is a clinically important problem for young women with early breast cancer, leading to distressing symptoms such as hot flashes, loss of fertility and increased risk of bone loss [7]. Five published surveys of menopausal symptoms in women with breast cancer have shown their importance to patients [8–12]. Three of these studies included both pre- and postmenopausal women, and all of them included data from breast cancer survivors who had not received adjuvant chemotherapy: thus, they probably underestimated the frequency and severity of symptoms in those where menopause is induced by chemotherapy. Two studies have compared menopausal symptoms in women with breast cancer with a healthy control group [11, 12], but the groups in these studies were not matched for either age or menopausal status. In the present study, patients who had undergone a chemotherapy-induced menopause were well matched by time since last menstrual period with control women recruited largely from a breast-screening program. The results support our hypothesis that women undergoing chemotherapy-induced menopause experience more frequent and severe symptoms than women undergoing natural menopause. There was an effect of borderline significance for the patient population to experience worse symptoms (as determined by the median FACT-ES score) when compared with a group of healthy women, but there was a substantial difference in patients who experienced moderate–severe hot flashes; these results indicate that symptoms vary among women with chemotherapyinduced menopause (hence small differences in median scores) but that a substantial minority experience worse symptoms. This might occur because the process of menopause, which occurs naturally over several years, is compressed into weeks or months. A striking finding of our previous study was the profound fatigue reported by women during and after chemotherapy that was dramatically worse than in age-matched controls [2, 3]. In

986 | Mar Fan et al.

our current study, there was a strong trend toward more fatigue in patients than in controls, despite the comparison being made when the patients had long since completed chemotherapy. We may have underestimated fatigue and overestimated quality of life, particularly in patients, as there is adaptation leading to a response shift for evaluating symptoms and quality of life with questionnaires so that the frame of comparison becomes different for patients and controls [13]. An association between fatigue and menopausal symptoms was observed in both patient and control groups in the present study: our analysis indicates that while menopause contributes to fatigue (perhaps by nocturnal vasomotor symptoms disrupting sleep [14]), it is not the only causative factor. Several factors may contribute to menopausal symptoms experienced by breast cancer survivors. These include the relative contraindication of HRT, lack of safe and effective alternatives to HRT, and the use of adjuvant hormonal therapy as part of the systemic management of early breast cancer. Since the publication of data questioning the safety of HRT, its routine use by the general population has declined [15], and few controls were receiving HRT in the current study. Approximately half the patients in our study were receiving adjuvant hormonal therapy for breast cancer, and others have reported that such treatments increase the incidence of menopausal symptoms, especially in women who are already postmenopausal [16, 17]. Rather surprisingly, our multivariate analysis failed to identify an association between adjuvant hormonal therapy and severity of menopausal symptoms; this may be due to limited power of the study and/or to chemotherapy having the dominant effect to reduce levels of estrogen and induce menopausal symptoms in younger women [2, 3]. The risk of chemotherapy-induced menopause is age related, at least for patients who receive mainly combination chemotherapy with cyclophosphamide, methotrexate and fluorouracil [1]. We carried out an unplanned analysis of our current data to see if the same trend applies to more modern chemotherapy regimens and also observed increasing risk for menopause with increasing age of the patient. Studies of this kind face methodological challenges. It is not possible to match controls for both age and menopausal status. It was not possible to recruit a control group of menopausal breast cancer patients who had not received either chemotherapy or hormonal therapy because most women with breast cancer receive adjuvant treatment. Nonhormonal drugs, such as SSRI (e.g. venlafaxine) and gabapentin, are used for treatment of hot flashes [18]; including patients who were

Volume 21 | No. 5 | May 2010

Downloaded from http://annonc.oxfordjournals.org/ at Northern Arizona University on June 9, 2015

(FACT-F score, P = 0.05 and 0.006, respectively). Poorer overall quality of life was associated with fatigue for both patients and controls (P <0.0001 and P = 0.03, respectively). No other associations were seen. Differences in scores of the FACT-ES, FACT-G and FACT-F between the two time points were not significantly different for either patients or controls.

original article

Annals of Oncology

acknowledgement We thank the OBSP for the assistance they provided in conducting this study.

Volume 21 | No. 5 | May 2010

references 1. Goodwin PJ, Ennis M, Pritchard KI et al. Risk of menopause during the first year after breast cancer diagnosis. J Clin Oncol 1999; 17: 2365–2370. 2. Tchen N, Juffs HG, Downie FP et al. Cognitive function, fatigue, and menopausal symptoms in women receiving adjuvant chemotherapy for breast cancer. J Clin Oncol 2003; 21: 4175–4183. 3. Mar Fan HG, Houe´de´-Tchen N, Yi QL et al. Fatigue, menopausal symptoms, and cognitive function in women after adjuvant chemotherapy for breast cancer: 1and 2-year follow-up of a prospective controlled study. J Clin Oncol 2005; 23: 8025–8032. 4. Fallowfield LJ, Leaity SK, Howell A et al. 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: 189–199. 5. Yellen SB, Cella DF, Webster K et al. Measuring fatigue and other anemia-related symptoms with the Functional Assessment of Cancer Therapy (FACT) measurement system. J Pain Symptom Manage 1997; 13: 63–74. 6. 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: 570–579. 7. Shapiro C, Manola J, Leboff M. Ovarian failure after adjuvant chemotherapy is associated with rapid bone loss in women with early stage breast cancer. J Clin Oncol 2001; 19: 3306–3311. 8. Canney PA, Hatton MQ. The prevalence of menopausal symptoms in patients treated for breast cancer. Clin Oncol (R Coll Radiol) 1994; 6: 297–299. 9. Couzi RJ, Helzlsouer KJ, Fetting JH. Prevalence of menopausal symptoms among women with a history of breast cancer and attitudes toward estrogen replacement therapy. J Clin Oncol 1995; 13: 2737–2744. 10. Carpenter JS, Andrykowski MA. Menopausal symptoms in breast cancer survivors. Oncol Nurs Forum 1999; 26: 1311–1317. 11. Harris PF, Remington PL, Trantham-Dietz A et al. Prevalence and treatment of menopausal symptoms among breast cancer survivors. J Pain Symptom Manage 2002; 23: 501–509. 12. Conde DM, Pinto-Neto AM, Cabello C et al. Menopause symptoms and quality of life in women aged 45 to 65 years with and without breast cancer. Menopause 2005; 12: 436–443. 13. Sprangers MA, Schwartz CE. The challenge of response shift for quality-of-lifebased clinical oncology research. Ann Oncol 1999; 10: 747–749. 14. Glaus A, Boehme Ch, Thurlimann B et al. Fatigue and menopausal symptoms in women with breast cancer undergoing hormonal cancer treatment. Ann Oncol 2006; 17: 801–806. 15. Hickey M, Davis S, Sturdee D. Treatment of menopausal symptoms: what shall we do now? Lancet 2005; 366: 409–421. 16. Harris PF, Remington PL, Trentham-Dietz A et al. Prevalence and treatment of menopausal symptoms among breast cancer survivors. J Pain Symptom Manage 2002; 23: 501–509. 17. Morales L, Neven P, Timmerman D et al. Acute effects of tamoxifen and thirdgeneration aromatase inhibitors on menopausal symptoms of breast cancer patients. Anticancer Drugs 2004; 15: 753–760. 18. Bordeleau L, Pritchard K, Goodwin P et al. Therapeutic options for the management of hot flashes in breast cancer survivors: an evidence-based review. Clin Ther 2007; 29: 230–241.

doi:10.1093/annonc/mdp394 | 987

Downloaded from http://annonc.oxfordjournals.org/ at Northern Arizona University on June 9, 2015

receiving treatment for hot flashes may have biased the results, but excluding them would have created more bias, and we addressed this by means of multivariate analysis. Additional factors, which might impact on menopausal symptoms, such as smoking and body mass index, were not assessed. Recruitment of healthy women who had recently gone through menopause proved to be challenging with almost 3000 women screened to achieve our sample size of 57. The study did not achieve its original planned sample size but a recruitment strategy that targeted women seeking management of menopausal symptoms would result in a biased cohort of more symptomatic women. The study is underpowered to detect small differences between groups and caution should be exercised in the generalization of its results. Our study did succeed in recruiting a control population, which is very well matched to the patient group in terms of median duration since last menstrual period, allowing comparisons to be made between the two groups. Perhaps the ideal study design for comparing premature and natural menopause would be to follow two cohorts of women, one group approaching natural menopause and another at risk of chemotherapy-induced menopause. This would require a large initial sample size and multiple assessments over a number of years and was beyond the scope of our current study. Despite the aforementioned problems, the present comparison of normal and chemotherapy-induced menopause has provided important information and is the first to specifically make this comparison. An informed decision about adjuvant chemotherapy for breast cancer should include knowledge of the risk of premature menopause and the frequency and severity of the associated symptoms. Identification of menopausal symptoms will allow appropriate interventions to be made. Extrapolation of experience from natural menopause to patients with chemotherapy-induced menopause is inappropriate given the different severity of symptoms and the available treatment options. Given the recent decrease in mortality from breast cancer and the increasing number of survivors, long-term side-effects of adjuvant chemotherapy are increasingly important.