Fax: (617) 632-1930
The effects of paclitaxel, dose density, and trastuzumab on treatment-related amenorrhea in premenopausal women with breast cancer
Article first published online: 5 JAN 2010
Copyright © 2010 American Cancer Society
Volume 116, Issue 4, pages 791–798, 15 February 2010
How to Cite
Abusief, M. E., Missmer, S. A., Ginsburg, E. S., Weeks, J. C. and Partridge, A. H. (2010), The effects of paclitaxel, dose density, and trastuzumab on treatment-related amenorrhea in premenopausal women with breast cancer. Cancer, 116: 791–798. doi: 10.1002/cncr.24835
- Issue published online: 2 FEB 2010
- Article first published online: 5 JAN 2010
- Manuscript Accepted: 15 JUN 2009
- Manuscript Revised: 21 MAY 2009
- Manuscript Received: 16 FEB 2009
- dose-dense regimens;
- menopausal status;
Little information is available regarding the effects of new adjuvant treatment regimens on menstrual functioning in premenopausal women with early breast cancer.
The authors conducted a retrospective review of data from premenopausal women who received treatment for early breast cancer to evaluate the rates of amenorrhea in follow-up. The women who were included received treatment with either doxorubicin and cyclophosphamide (AC) or combined AC and paclitaxel (T) (AC-T) given either every 3 weeks, or as a dose-dense (DD) regimen, or as AC followed by weekly T with trastuzumab or followed by trastuzumab (AC-T+trastuzumab). A multivariate logistic regression analysis was conducted to evaluate amenorrhea during follow-up.
Of 431 patients who were eligible for analysis, the average age at diagnosis was 13 years (range, 25-55 years), 61% of women received AC only, and 39% received AC-T. Of the 39% who received AC-T, 49% of women received DD therapy, 14% received AC-T+trastuzumab, and 71% of all patients received tamoxifen (TAM). The median follow-up was 33 months (range, 6-114 months). After adjusting for age, weight, gravidity, parity, age at menarche, smoking, alcohol use, TAM use, type and regimen of chemotherapy, and use of trastuzumab, the likelihood of remaining amenorrheic was not statistically different in patients who received AC-T versus AC (odds ratio [OR], 1.59; 95% confidence interval [CI], 0.8-3.2), DD treatment versus treatment every 3 weeks (OR, 0.56; 95% CI, 0.25-1.3), or AC-T + trastuzumab (OR, 0.6; 95% CI, 0.22-1.61). Amenorrhea was associated significantly with TAM use and age at diagnosis.
Recent advances in the adjuvant treatment of early breast cancer do not appear to have increased the risk of amenorrhea in premenopausal women. Cancer 2010. © 2010 American Cancer Society.
Of the estimated 178,480 new diagnoses of breast cancer in 2007, approximately 25% occurred in premenopausal women.1, 2 Earlier diagnosis and treatment advances, including the widespread use of chemotherapy in premenopausal women, have been successful in increasing the 5-year survival rate among young patients with breast cancer.3 Consequently, breast cancer survivors are living longer with the side effects of their treatment. For many young women who are surviving cancer, few issues are more relevant than that of the impact of treatment on their reproductive function.
Adjuvant chemotherapy may render a premenopausal woman amenorrheic, either temporarily or permanently.4-7 In women with a history of hormone receptor-positive breast cancer, premature menopause has been associated with a reduction in the risk of recurrence; however, it is not clear that this adds to the benefits of the current standard adjuvant tamoxifen (TAM) therapy. Premature menopause clearly can have profoundly deleterious effects on quality of life. Physical ramifications of menopause are multifold and include cardiovascular system changes, osteoporosis, altered sexual function, and hot flushes. Early menopause also truncates the reproductive potential of survivors, and facing a diagnosis of infertility after breast cancer can be challenging. Without question, the impact of chemotherapy on ovarian function can have an adverse effect on a woman's emotional well being8 and quality of life on a long-term basis. For some women who are facing a diagnosis and treatment of breast cancer, fertility concerns may be so powerful that they have an impact on treatment decisions.9 Furthermore, understanding a woman's risk of becoming postmenopausal after chemotherapy has important implications for future treatment (eg, consideration of adjuvant aromatase-inhibitor therapy) and family planning (eg, the need for contraception).
The link between adjuvant chemotherapy and amenorrhea has been well established, but the specific effects of individual chemotherapeutic agents and regimens have not been well characterized. In particular, data pertaining to the use of taxanes are sparse and controversial. Similarly, almost no information exists on the use of a dose-dense (DD) schedule (treatment every 2 weeks) on chemotherapy-related amenorrhea (CRA). In addition, the impact of trastuzumab on CRA is not well documented. The primary objectives of the current study were to evaluate the relation between CRA at a follow-up ≥6 months and the addition of paclitaxel (T) to doxorubicin and cyclophosphamide (AC) and to explore the effects of the use of a DD regimen and the use of trastuzumab, with or without TAM, in premenopausal women with breast cancer.
MATERIALS AND METHODS
We identified patients by using an institutional database for which data are input prospectively for all women who undergo breast cancer treatment at Dana-Farber Cancer Institute and Massachusetts General Hospital. Informed consent was obtained from patients before entry of their medical information and personal characteristics into the database. The Institutional Review Board of Dana Farber/Harvard Cancer Center approved this study. Eligible women had been diagnosed with early stage breast cancer and received adjuvant chemotherapy from 1997 through 2005. Women were selected for this analysis from the institutional database if they received adjuvant AC chemotherapy (doxorubicin 60 mg/m2 and cyclophosphamide 600 mg/m2) either every 3 weeks or as DD therapy, with or without the addition of T or T + trastuzumab, and if they were premenopausal at diagnosis. Women were classified as premenopausal in the database if they had menses in the past 6 months and were classified as postmenopausal if they had no menses in the past 6 months because of natural menopause; removal of both ovaries; chemotherapy, radiation therapy, or endocrine therapy for a prior nonbreast cancer; ovarian failure from a medical condition; or had a menstrual period in the past 6 months during hormone-replacement therapy. This information was obtained from the baseline patient survey with back-up chart review before it was input into the database. Patients with contradictory or missing data were classified as postmenopausal if they were aged ≥50 years. For this study, a detailed chart review was used to confirm eligibility regarding premenopausal status at diagnosis and the receipt of 1 of the chemotherapy regimens of interest. The chemotherapy regimens allowed included: 4 cycles of AC only, AC followed by 4 cycles of T (paclitaxel 175 mg/m2) (AC-T), or AC followed or preceded by 1) weekly T (paclitaxel 80 mg/m2 with weekly trastuzumab [2 mg/kg, with an initial loading dose of 4 mg/kg] for 12 weeks, followed for some patients by 40 additional weeks of trastuzumab after the completion of conventional cytotoxic chemotherapy), or 2) weekly T alone × 12 followed by weekly trastuzumab for 52 weeks (grouped as AC-T + trastuzumab). Information regarding the receipt of TAM (20 mg daily) also was obtained from chart review. For the primary objectives of this study, all women were required to have menstrual function information in the medical record for a minimum of 6 months after treatment cessation. Patients were excluded if they had undergone surgical or medical ovarian ablation early in follow-up, if they had received prior treatment with chemotherapy, if there was no follow-up, or if their menstrual status could not be ascertained in follow-up from the medical record. Patients who developed recurrences that required additional chemotherapy were included in the analysis up until the time of their new treatment.
For this analysis, we defined “amenorrhea” as the continued absence of menstrual periods as recorded in the medical record at time intervals of 6 months, 6 to 12 months, and >12 months from the date of the patient's last dose of chemotherapy with AC or T with or without trastuzumab. If a patient reported the presence of cyclic bleeding at any point in an interval, then she was not considered amenorrheic for that interval. Patients without evidence of cyclic bleeding within the interval of interest were considered amenorrheic. For this analysis, a single, isolated episode of bleeding was not counted as a resumption of menstrual function and, instead, was assumed to be similar to an episode of “postmenopausal bleeding.”
The distribution of patient characteristics was quantified among all women and stratified by amenorrheic status throughout follow-up. Two models were fit to the data to identify relations between chemotherapy regimen and the likelihood of remaining amenorrheic. Unconditional multivariate logistic regression—adjusted for age at diagnosis, age at menarche, weight, gravidity, parity, smoking, alcohol consumption, type and regimen of chemotherapy, TAM use, trastuzumab use and length of follow-up—was used to calculate odds ratios (ORs) and 95% confidence intervals (CIs), which are presented as estimates of the relative risk. To evaluate heterogeneous associations among women who were older versus those who were younger at diagnosis, we also applied these models to the study population stratified at age 40 years.
All multivariate Wald statistic P values are based on 2-sided tests. The SAS statistical software package (version 9.1; SAS Institute Inc., Cary, NC) was used for all analyses.
We identified 431 premenopausal breast cancer patients who met eligibility criteria. (Table 1) Eight hundred ninety-two premenopausal breast cancer patients were identified originally from the institutional database who received AC chemotherapy either every 3 weeks or on a DD schedule, with or without the addition of T or T + trastuzumab, However, after a detailed chart review, 461 patients were excluded based on the following reasons: Three hundred forty-six patients received chemotherapy with drugs or regimens different from the study criteria, 64 patients had undergone hysterectomy and/or removal of bilateral ovaries before diagnosis, 10 patients underwent hysterectomy after completing chemotherapy but before 6 months of follow-up, 26 patients had insufficient information about menstrual status after follow-up, 14 patients received luprolide or aromatase inhibitors during and/or after chemotherapy, and 1 patient had undergone uterine artery embolization before diagnosis. The median follow-up was 33 months (range, 6-114 months). Data were available for 100% of patients at the 6-month interval after the completion of chemotherapy, for 97% of patients at the 6-month to 12-month interval, and for 96% of patients at the >12-month interval. One hundred ninety-two women (45%) remained premenopausal after completion of chemotherapy, and 239 women (55%) reported no menses during follow-up. Among the women who reported menses, 119 of 192 (62%) reported menstruating within 6 months of treatment cessation, 60 of 192 (31%) reported menstruating within 12 months, and an additional 13 of 192 (7%) reported menstruating >12 months after cessation. (Table 2) Hence, if menses were to be present after the completion of treatment, the majority of women reported menstruating within the first 6-month interval, whereas the proportion that experienced the return of menses >12 months after treatment cessation was small.
|Variable||Total Study Population, N=431|
|No. of Patients (%)||Median [Range]|
|Total follow-up, mo||431||33 [6-114]|
|Age at diagnosis, y||431||43 [25-55.6]|
|Age at menarche, y||427||13 [9-18]|
|Age at first birth, y||310||29 [15-45]|
|Weight, kg||415||65 [42.3-119]|
|Height, inches||373||163 [145-180]|
|BMI, kg/m2||370||23.9 [15.1-46.6]|
|Alcohol, drinks per wk|
|Underweight: <18.5 kg/m2||47 (12.7)|
|Overweight/obese: >25.0 kg/m2||145 (39.2)|
|AC and paclitaxel|
|4 Doses||170 (39.4)|
|12 Doses||33 (7.7)|
|AC, paclitaxel, and trastuzumab|
|12 Doses||15 (3.5)|
|52 Doses||24 (5.6)|
|Dose density||120 (27.8)|
|Time of Reported Presence of Menses, mo||No. of Patients (%)|
|From 12 to end of follow up||13 (3)|
|Amenorrheic at end of follow up||239 (55)|
After adjusting for age, weight, gravidity, parity, age at menarche, smoking, alcohol use, TAM use, type and regimen of chemotherapy, use of trastuzumab, and length of follow-up, the multivariate model revealed that the likelihood of remaining amenorrheic during follow-up was associated significantly with TAM use and age at diagnosis (Table 3). Patients who were receiving TAM therapy were twice as likely to remain amenorrheic compared with those who were not receiving TAM (OR, 2.12; 95% CI, 1.13-4.00). Similarly, increasing age at diagnosis was associated positively with an increased risk of developing CRA (test for trend; P < .0001). For example, patients who were ages 35 years to 39 years at diagnosis were 10 times as likely to remain amenorrheic compared with patients who were aged <35 years at diagnosis (OR, 10.1; 95% CI, 1.28-79.0). The likelihood of remaining amenorrheic did not differ statistically for patients who received AC-T versus AC (OR, 1.59; 95% CI, 0.80-3.22), patients who received DD treatment versus treatment every 3 weeks (OR, 0.56; 95% CI, 0.25-1.3), or patients who received trastuzumab in addition to chemotherapy (OR, 0.6; 95% CI, 0.22-1.61). These relations were similar to those observed in unadjusted analyses, as noted in Table 3, and, thus, were robust to confounding by factors that were included in the multivariate models.
|Variable||Unadjusted Model||Multivariate Modela|
|OR||95% CI||P||OR||95% CI||P|
|Age at diagnosis, y|
|≥45||604.9||77.2 to >999||<.0001||558.7||70.6 to >999||<.0001|
|P test for trend||<.0001||<.0001|
When stratifying by age at diagnosis and evaluating the relation between specific treatment regimens and the likelihood of remaining amenorrheic, we observed that, among patients who were aged ≥40 years at diagnosis (N = 296), TAM use was associated with a 2-fold increased likelihood of remaining amenorrheic (OR, 2.51; 95% CI, 1.20-5.24) (Table 4). However, among women who were aged <40 years at diagnosis (N = 138), the significant association with TAM was not observed (OR, 1.89; 95% CI, 0.52-6.89). No statistically significant difference in the likelihood of CRA was observed among patients who were ages <40 years or >40 years at diagnosis who received treatment with or without T or trastuzumab or those who received their chemotherapy according to a DD schedule versus a schedule of every 3 weeks.
|Variable||Age <40 Years at Dx (N=135)b||Age ≥40 Years at Dx (N=296)b|
|OR||95% CI||P||OR||95% CI||P|
An assessment of women with treatment-related amenorrhea by age and chemotherapy revealed that, regardless of the specific regimen used, women aged <40 years had a much lower incidence of amenorrhea after treatment than women between ages 40 years and 49 years (Table 5). From 94% to 100% of women who were aged >50 years at the onset of treatment became amenorrheic after treatment, whereas only 0% to 16% of women aged <40 years were amenorrheic after treatment. The small number of women in certain subgroups limited the results.
|Treatmentb||No. of Amenorrheic Women/Total No. (%)|
|Aged <40 Years, N=18/135 (13.3%)||Ages 40-49 Years, N=172/246 (69.9%)||Aged ≥50 Years, N=49/50 (98%)|
|AC||10/72 (13.9)||88/129 (68.2)||27/27 (100)|
|AC and paclitaxel|
|4 Doses||7/52 (13.5)||72/98 (73.5)||20/21 (95.2)|
|12 Doses||1/11 (9.1)||13/20 (65)||2/2 (100)|
|AC, paclitaxel, and trastuzumab|
|12 Doses||0/6 (0)||5/9 (55.6)||0 (0)|
|52 Doses||1/7 (14.3)||10/15 (66.7)||2/2 (100)|
|Dose-dense regimen||5/36 (13.9)||49/67 (73.1)||17/18 (94.4)|
|Tamoxifen||13/80 (16.3)||135/183 (73.8)||40/41 (97.6)|
In this evaluation of follow-up menstrual functioning in premenopausal women with invasive breast cancer who received adjuvant chemotherapy with AC with or without the addition of T, trastuzumab, or TAM, given either every 3 weeks or as a DD regimen, we observed no statistically significant increased likelihood of CRA in patients who received T or trastuzumab compared with those who received AC alone. Similarly, there was no statistically significant increase in the likelihood of CRA in patients who were treated on DD regimens compared with those who were treated on standard regimens (every 3 weeks). Although there may be a trend toward an effect, any undetected, small increase in the risk of CRA is unlikely to have substantial clinical implications. Our findings suggest that 2 factors significantly increased the likelihood of CRA: age at diagnosis and TAM use. This significant association of TAM and age on the risk of CRA is consistent with previous studies that quantified the risk of CRA with anthracycline-based adjuvant chemotherapy.10, 11 Our findings support those reported in previous studies: For women aged <30 years, premature ovarian failure with standard regimens is very uncommon. The observed risk of CRA with 4 cycles of standard AC was <1% among women aged <30 years.5, 12, 13 With 4 cycles of AC, the risk of premature ovarian failure increases up to the 10% to 15% range among all women aged <40 years.13-15 The risk in younger women after 6 cycles of combined cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) or combined cyclophosphamide, epirubicin, and 5-fluorouracil is somewhat higher; with these regimens, up to 20% of women aged ≤30 years and up to 40% of women aged ≤40 years will experience premature menopause.11, 13-15 Age at the onset of adjuvant chemotherapy was a significant linear predictor of CRA. Compared with women who were aged <25 years at diagnosis, women who were between ages 35 years and 39 years at diagnosis were 10 times more likely to remain amenorrheic. Women who were between ages 40 years and 44 years at diagnosis were nearly 40 times more likely to remain amenorrheic, and women who were aged >45 years at diagnosis were >500 times more likely to remain amenorrheic.
It has been reported that TAM improves disease-free survival in premenopausal women who have positive estrogen receptor status.16 The relation of TAM and CRA is unclear in the literature. Some studies report no increase the risk of CRA for patients who take TAM,17-19 whereas others report a higher incidence of CRA among patients who use TAM.4, 16, 20, 21 These inconsistencies may reflect differences in the ages of women in the studies as well as the length of follow-up. Our study suggests that the effect of TAM on CRA varies by age, because women aged ≥40 years were more likely to have CRA when they were taking TAM, whereas TAM appeared to have less of an association with CRA in younger women. Although TAM clearly is related to menstrual irregularities in some women while they are on the medication, there is no evidence nor is it biologically probable that TAM directly damages the ovaries. However, women would be expected to be less fertile after TAM or other hormone treatment because of the natural decline of ovarian function over the time it takes to receive standard therapy.
To our knowledge, this is the largest study to date that is focused on the effect of adding T to AC chemotherapy on subsequent menstrual function. Although some previous studies reported an increased risk of amenorrhea with the addition of taxanes, including T19, 22-26 or docetaxel (D),22, 25, 26 others reported a decreased or equivalent risk with T and/or D or with T alone.23, 27, 28 In a retrospective analysis in which 60 patients received AC alone and 21 patients received AC + T on a standard dosing regimen, CRA was less likely in the group that received T.27 In a prospective study of 64 women, 32 women received AC alone, whereas 14 women received AC followed by T on a standard dosing regimen, Alton et al observed that women who received AC followed by T were 33% as likely to experience CRA as women who received AC only.29 In another retrospective study, Fornier et al demonstrated no change in the effect on menstrual functioning with the addition of taxanes, T, and/or D.18 Wide variation in the specific agents (including AC, CMF, T, or D) and doses used, as well as the definition of amenorrhea, may account in part for these differences. Findings from the current study in which the risk of CRA was evaluated in patients who received AC with or without the addition of T on standard versus DD regimens did not demonstrate a statistically significant increased risk of CRA in patients who received T on either standard or DD regimens. This study adds to the evidence that the addition of T does not appreciably increase the risk of CRA. Although the administration of AC and T in a DD regimen reportedly is not accompanied by substantially increased conventional chemotherapy toxicity,30 little previous information is available to assess the risk of CRA. In 1 previous analysis, patients received equal doses of epirubicin given every 14 days or 21 days, and the incidence of amenorrhea was equal in both arms.31 In the study by Fornier et al, 34 of 166 patients who were evaluated had received AC followed by T on a DD regimen, and no apparent increase in the risk of amenorrhea was observed in these patients.18 Our findings confirm these results.
To our knowledge, the risk of CRA related to treatment with trastuzumab, has not been assessed previously. Almost 25% of breast cancer patients have tumors that over express human epidermal growth factor receptor 2 (HER2).32 Today, trastuzumab is used widely for the treatment of these tumors based on results from several large, randomized trials.20, 33 In the current study, we observed no increase in the likelihood of CRA for patients receiving trastuzumab on short-term (12 weeks) or long-term (52 weeks) regimens. Although we were limited by the small sample size and diversity of regimens used, our preliminary findings may be useful for patients who are receiving trastuzumab for HER2-positive breast cancer, and attempts to replicate this observation should be undertaken in larger studies.
Limitations of the current study include its retrospective design and the small numbers for certain subgroups (eg, trastuzumab). Information on baseline menstrual functioning was obtained through the retrospective review of prospectively collected data and was confirmed by a detailed chart review. Women under the stress of cancer treatment may be less focused on their menstruation status at diagnosis and in follow-up. Similarly, women who are postchemotherapy and concerned about fertility may be more apt to remember even the subtlest details of their cycles. Furthermore, missing information or potentially inaccurate follow-up menstrual information recorded by providers in the medical record may introduce bias. The large sample size of the current study, compared with the relatively small number of women who were excluded from this analysis for missing menstrual data in the follow-up data, is reassuring. In ongoing work, prospectively collected data obtained by patient interview or questionnaires specifically geared toward menstrual functioning and not dependent on patient or physician recall will limit these biases.
Our findings may assist in the counseling of premenopausal women with breast cancer. Treatments that may be viewed as more aggressive, such as the addition of taxanes and trastuzumab and the use of DD regimens, do not necessarily lead to an increased risk of CRA, a dreaded consequence of chemotherapy for some premenopausal women. Data from our study demonstrate that a significant proportion of premenopausal women who are diagnosed with breast cancer will continue to menstruate for at least some time after chemotherapy. Clearly, however, menstruation is an imperfect surrogate for fertility, and ovarian reserve markers may reflect more accurately the damaging effects of cytotoxic therapy on ovarian function. Recent research has revealed that even women who continue to menstruate after chemotherapy may have diminished markers of ovarian reserve in follow-up.34, 35 Further research in this population is warranted to determine ovarian reserve and, ultimately, fertility outcomes and to determine the timing of menopause in survivors who remain premenopausal after modern breast cancer chemotherapy regimens.
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
- 1RiesL, EisnerM, KosaryC, et al. eds. SEER Cancer Statistics Review, vol 2007, 1975-2001. Bethesda, Md: National Cancer Institute; 2005.
- 10Chemotherapy-induced amenorrhea and fertility in women undergoing adjuvant treatment for breast cancer. Cancer Control. 2002; 9: 466-472., .
- 11Reproductive issues. In: HarrisJR, ed. Diseases of the Breast. 2nd ed. Philadelphia, Pa: Lippincott Williams & Wilkins; 2000: 1051-1059., .
- 13Immediate and long-term toxicity of adjuvant chemotherapy regimens containing doxorubicin in trials at M.D. Anderson Hospital and Tumor Institute. NCI Monogr. 1986; 1: 105-109., , , .
- 19Chemotherapy-induced amenorrhea is increased in patients treated with adjuvant doxorubicin and cyclophosphamide (AC) followed by a taxane (T). 2005 ASCO Annual Meeting Proceedings [abstract]. J Clin Oncol. 2005; 23( part I of II; June 1 suppl): 16S. Abstract651., , , , .
- 23The impact of paclitaxel on menstrual function. Breast Cancer Res Treat 2005; 94( suppl): 271s-272s., , , et al.
- 25Safety analysis of the PACS 01 adjuvant trial comparing 6 cycles of FEC 100 followed by 3 cycles of docetaxel (Taxotere) for node positive breast cancer [abstract]. Breast Cancer Res Treat. 2003; 82. S32. Abstract 144., , , et al.
- 26Chemotherapy-induced amenorrhea (CIA) in patients treated with adjuvant CEF/CMF or EC/docetaxel: analysis from a phase III randomized EC/Doc Trial [abstract]. Breast Cancer Res Treat. 1005; 94( S105). Abstract 2063., , .
- 27Rate of chemotherapy related amenorrhea (CRA) associated with adjuvant Adriamycin and Cytoxan (AC) and Adriamycin and Cytoxan followed by Taxol (AC+T) in early stage breast cancer [abstract]. Breast Cancer Res Treat. 2000; 64: 61. Abstract 224., , , et al.
- 28Effect of paclitaxel (P)-based chemotherapy on the ovarian failure (OF) of breast cancer patients (pts): a retrospective study [abstract]. Proc Am Soc Clin Oncol. 2003; 22. Abstract 3029., , , et al.
- 29Chemotherapy-related amenorrhea (CRA) in breast cancer survivors: impact of taxanes on ovarian function [abstract]. Breast Cancer Res Treat. 2004; 88( suppl 1): 61s. Abstract 1060., , , et al.
- 30Randomized trial of dose-dense versus conventionally scheduled and sequential versus concurrent combination chemotherapy as postoperative adjuvant treatment of node-positive primary breast cancer: first report of Intergroup Trial C9741/Cancer and Leukemia Group B Trial 9741. J Clin Oncol. 2003; 21: 1431-1439., , , et al.