Chemotherapy-induced amenorrhea: influence on disease-free survival and overall survival in receptor-positive premenopausal early breast cancer patients

M. Vanhuyse, C. Fournier and J. Bonneterre*

Centre Oscar Lambret, Lille, France

* Correspondence to: Dr J. Bonneterre, Département de Senologie, Centre Oscar Lambret, 3 Rue Frédéric Combemale, 59020 Lille Cedex, France. Tel: +33-3-20-29-55-50; Fax: +33-3-20-29-55-80; Email: j-bonneterre{at}o-lambret.fr


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background:: The aim of this study was to evaluate the influence of early chemotherapy-induced amenorrhea (CIA) on disease-free survival and overall survival in premenopausal patients with receptor-positive early breast cancer treated with adjuvant chemotherapy without any hormonotherapy.

Patients and methods:: Retrospectively, we reviewed data from 130 premenopausal patients with localized hormone-sensitive breast cancer. These patients were treated between 1985 and 1995 at the same institution. They all underwent a loco-regional treatment and adjuvant chemotherapy. Early CIA was defined as an amenorrhea arising during the first year following the beginning of chemotherapy. Predictors of early CIA were examined. The survival analyses were done using the Kaplan–Meier method and Cox analysis.

Results:: Median follow-up was 9 years. Mean age was 42.9 ± 5 years. Ninety-two per cent of patients had histologically-proven positive axillary nodes. Adjuvant chemotherapy contained no anthracycline in 63%. Early CIA occurred during or after adjuvant chemotherapy in 57% of the patients. It was definitive in 91%. In our study, age was the only CIA predictor in univariate analysis. Women who experienced early CIA tend to have a longer disease-free survival, but the difference was not significant. This trend was lost in multivariate analysis, most probably due to the small sample size. The overall survival was not different.

Conclusion:: Although not statistically significant, our results on a very selected population of patients suggest that a chemotherapy-induced amenorrhea might have its own therapeutic effect besides the cytotoxic action of chemotherapy.

Key words: amenorrhea, adjuvant chemotherapy, breast cancer


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Analyses of randomized clinical trials have clearly demonstrated a relapse and mortality reduction by using adjuvant chemotherapy and hormonotherapy in early breast cancer [1Go, 2Go]. The advantage derived from adjuvant chemotherapy seems to be higher in premenopausal women. Moreover, chemotherapy can induce amenorrhea, which is often definitive [3Go]. This chemotherapy-induced castration can be partly responsible for the action of chemotherapy in premenopausal patients, over direct cytotoxicity. The chemotherapy-induced amenorrhea (CIA) has been evaluated as a prognostic factor in some studies [4Go–7Go]. These studies were always retrospective and included heterogeneous populations, which mixed hormone-sensitive and resistant tumors as well as hormonotherapy-treated and untreated patients. Moreover, amenorrhea was not consistently defined and permanent and temporary amenorrheas were not differentiated.

The aim of this study was to analyze the influence of early CIA on disease-free and overall survival in a targeted population. Premenopausal women treated with adjuvant chemotherapy for a receptor-positive breast cancer are considered. We studied the early CIA rather than late CIA, which was more likely to be a physiological amenorrhea rather than a chemotherapy-induced amenorrhea [8Go].


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Study design
In this retrospective study, we reviewed data from 242 patients treated in the Oscar Lambret Center, the Comprehensive Cancer Center of Northern France, between 1985 and 1995. This period was chosen for two reasons: first, to have a long duration of follow-up; and secondly, because at this time, no adjuvant hormonotherapy was routinely given in premenopausal patients.

Definitions
Premenopausal status was clinically defined by regular menstruation at initial diagnosis. For the early CIA, we used the Bines' definition [3Go]. Early CIA was ≥6 months without menstrual periods and within 1 year after starting chemotherapy, in a patient who was premenopausal at diagnosis. Temporary CIA was the reappearance of regular menstruation after CIA had occurred.

Disease-free survival (DFS) was defined as the length of time from the date of the diagnosis to relapse (including ipsilateral breast recurrence), the appearance of a secondary primary cancer (ipsi or controlateral), or death. Overall survival (OS) was defined as the time between the diagnosis and the last follow-up or death.

Patients
All patients were premenopausal and presented a localized hormonosensitive breast cancer. Of the 242 patients, 112 were excluded for different reasons: (i) lack of information on the menstrual periods; (ii) previous history of bilateral oophorectomy or hysterectomy; (iii) adjuvant hormonotherapy as ovarian suppression (medical, surgical or by radiotherapy) or antioestrogen (tamoxifen); and (iv) non-specific breast cancer death, for example, one patient died of a sepsis during an aplasia. Finally, 130 patients were assessable.

Treatment
Surgery of the breast cancer was either a total mastectomy or a breast conserving procedure. An axillary lymph node dissection was always done. A loco-regional radiotherapy was delivered after surgery to all patients according to the protocol used at the center. The different protocols of adjuvant chemotherapy used were CMF-type (combination chemotherapy with cyclophosphamide, methotrexate and fluorouracil), FEC-type (50 or 100 mg/m2 epirubicin) and others [epirubicin-navelbine and peri-operative FAC (5-fluorouracil, doxorubicin, cyclophosphamide)] (Table 1). All patients received six cycles of chemotherapy except for peri-operative FAC with only one cycle. The patients never received adjuvant hormonotherapy.


View this table:
[in this window]
[in a new window]
 
Table 1. Patient characteristics

 
Statistical methods
First, univariate analysis was realized to determine the CIA predictors. Then, the Kaplan–Meier method was used to estimate survival distributions for DFS and OS in the presence or absence of CIA. A log-rank test was used to assess the statistical significance of differences between the DFS and OS distributions. The magnitude of differences in survival rates adjusted for covariates and the test for interactions were estimated with a Cox proportional hazards regression model. All tests were two-sided; a difference was considered significant at a P level of 0.05.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patient characteristics
The median duration of follow-up for censored patients was 9 years. Mean age was 42.9 ± 5 years. The patient characteristics are listed in Table 1. Overall, in this study, the women had a poor prognosis. In 42% of cases, more than three axillary lymph nodes were histologically involved. Adjuvant chemotherapy was realized for 10 women who had no node involvement because of age ≤35 years or because they were included in a clinical trial. Adjuvant chemotherapy in one-third of patients was an anthracycline-based regimen and was a variant of the CMF protocol in the remaining cases.

Univariate predictors of CIA
CIA occurred in 74 out of 130 (57%) patients. Most often, it happened during chemotherapy and was definitive in 91% of cases (Table 1). In univariate analysis, the incidence of CIA was significantly higher when women were older (Table 2). The mean age in the CIA group was 46.0 ± 4 years and in the no CIA group 38.8 ± 5 years (P <10–5). This result was well represented in the age distribution of patients shown in Figure 1. Among women who experienced a CIA, the temporary CIA was more frequent in younger patients (Figure 2) and occurred earlier (Figure 3). However, this difference was not statistically significant due to the small number of patients with temporary CIA (Table 1).


View this table:
[in this window]
[in a new window]
 
Table 2. Univariate analysis of CIA incidence according to clinical and tumor characteristics

 


View larger version (16K):
[in this window]
[in a new window]
 
Figure 1. Age distribution according to the CIA rate of the population.

 


View larger version (12K):
[in this window]
[in a new window]
 
Figure 2. Age distribution according to temporary or definitive CIA.

 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 3. Number of chemotherapy cycles before early CIA.

 
Other parameters were evaluated in univariate analysis as tumor characteristics (i.e. tumor size, histological type, nodal status, hormonal receptor and histopronostic grading) and treatment (i.e. hormonal treatment prior to diagnosis, surgical treatment, radiotherapy, chemotherapy regimen). None of them demonstrated a significant influence on the appearance of early CIA in our study (Table 2).

Disease-free and overall survival analysis
Half of the patients had a relapse, which was usually metastatic. Fifty-one patients died of breast cancer (Table 1). Relapse site and treatment were not statistically different between the CIA and no CIA groups. Nevertheless, the hepatic metastasis rate was higher in the no CIA group (34.4% versus 59.2%, P=0.10).

The Kaplan–Meier curves for DFS and OS according to CIA are shown in Figures 4 and 5. DFS was longer when a CIA occurred. However, the difference was not significant (P=0.11), even if the study is carried out for patients over and under 40 years (data not shown). Median DFS was 11.1 years in the presence of CIA and 7.2 years in its absence. OS curves, according to CIA status, were not different (P=0.26). Median OS was 13.3 years and 11.4 years in the presence and in the absence of CIA, respectively.



View larger version (14K):
[in this window]
[in a new window]
 
Figure 4. DFS according to CIA status.

 


View larger version (12K):
[in this window]
[in a new window]
 
Figure 5. OS according to CIA status.

 
When CIA, age, histological type, tumor size, histological grade and capsular node effraction were considered in a Cox multivariate model, no statistically significant difference was observed.


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The goal of this retrospective study was to evaluate the influence of early CIA on survival in a well-targeted population: premenopausal women treated with adjuvant chemotherapy for endocrine responsive breast cancer. We found a trend towards better disease-free survival when CIA occurred, which did not reach a statistically significant level. This result is consistent with the published data and the results are summarized in Table 3 [6Go, 9Go–22Go]. The retrospective studies were heterogeneous with regard to the number of patients, hormone receptor status and definitions of CIA and menopause. Most studies showed a significant improvement of DFS or OS when CIA occurred, particularly in hormone receptor-positive subpopulations. When the number of patients included in these studies was greater than 500, the significant statistical level was always reached. The power of our study was too low and probably means that the influence of CIA on DFS and OS is weak. Furthermore, in some studies, CIA was a prognostic factor in univariate but not in multivariate analysis [13Go, 17Go].


View this table:
[in this window]
[in a new window]
 
Table 3. Clinical studies on the effect of CIA as a prognostic factor

 
Unfortunately, the number of patients with temporary CIA was too low to study its influence on DFS and OS. Pagani, in his retrospective study, showed that temporary CIA was more frequent in younger women; furthermore, CIA did not need to be permanent to achieve a benefit on DFS [6Go].

Adjuvant chemotherapy, which improves survival in premenopausal women, provides a small hormonal effect by a chemo-induced castration and this effect was observed even if the CIA was temporary. A prospective study to confirm the results is no longer possible since premenopausal patients with hormonal receptor-positive tumors systematically receive adjuvant hormonotherapy.

In our experience, the only predictive factor of CIA is age. According to the literature many predictive factors can be added. For example, the chemotherapy regimen has an important effect on CIA incidence [3Go]. Furthermore, alkylating agents are more commonly associated with ovarian damages. The importance of ovarian damage can influence the duration of CIA. In our study, CIA incidence was 57%. It was 62% and 50% in the subgroups with CMF-type and FEC regimen, respectively. These data are consistent with the literature. Classical CMF regimens are associated with a 69% amenorrhea rate in a matched population [3Go]. Furthermore, cyclophosphamide cumulative dose and dose-density influence the CIA incidence. With anthracycline-based regimens, the CIA rate ranges from 33% to 59% [5Go]. Amenorrhea associated with taxanes is less well understood. Data from the BCIRG 01 trial comparing TAC and FAC in early-stage breast cancer show 51.4% and 32.8% CIA incidence, respectively [23Go]. In a recent publication, Di Cosimo found a relation between the incidence of CIA and the timing of treatment by menstrual cycle phase in women with early breast cancer treated with adjuvant chemotherapy [24Go].

Other methods induce castration: oophorectomy, radiotherapy and LHRH analogs. Many questions remain unanswered about the effectiveness of castration according to its duration in premenopausal women with endocrine-responsive tumors. Several data suggest little, if any, additional benefit from definitive castration (whatever the modalities) following adjuvant chemotherapy, suggesting that CIA would have a similar effect to castration. Oxford overviews have shown that adjuvant ovarian castration adds little compared to the effects of chemotherapy alone [25Go, 26Go]. The Davidson study compared CAF alone (CAF) versus CAF plus goserelin, LHRH analog, (CAFZ) versus CAF plus goserelin plus tamoxifen (CAFZT) [27Go]. The CAFZ arm produced a significantly better DFS and OS, whereas the addition of goserelin alone showed only marginal benefit. Unfortunately, this interesting study is not conclusive because of the lack of a CAF plus tamoxifen arm.

The role of castration instead of chemotherapy remains unclear [28Go]. Indeed, most trials that compared adjuvant ovarian ablation and adjuvant chemotherapy, used CMF regimens and not an anthracycline-based regimen except one, FAGS 06 [34Go]; chemotherapy regimen was often suboptimal, intravenous CMF rather than classical CMF, or FEC50 rather than FEC100. Currently, three clinical trials have been proposed by the International Breast Cancer Study Group (IBCSG) to investigate the relative role of hormonotherapy, including temporary or definitive ovarian suppression, and chemotherapy in premenopausal patients with hormonal receptor-positive early breast cancer. These studies, SOFT, TEXT and PERCHE, are summarized in Table 4.


View this table:
[in this window]
[in a new window]
 
Table 4. Design of the three IBCSG trials: SOFT, TEXT, PERCHE

 
While several studies analyzed the survival benefit of castration against chemotherapy, the different short and long-term side-effects of castration are poorly known. For example, Haes reported adverse effects observed in a prospective trial comparing six CMF cycles versus 2 years of goserelin in premenopausal women with node-positive breast cancer [32Go]. The results of this trial were published by Jonat [29Go, 30Go]. In endocrine responsive patients, both treatments had equivalent effects on DFS. Amenorrhea rates were different for patients on goserelin or CMF: 95.1% versus 58.6% at 6 months and 22.6% versus 76.9% at 3 years. The adverse effects related to oestrogen suppression (vaginal dryness and hot flushes) were greater during the 2 years of goserelin treatment, but were reversible. In another prospective study, Shapiro showed that women who experienced an CIA had rapid and important bone loss within 6 months of beginning chemotherapy [33Go]. In contrast, women who retained menstrual function had a non-significant decrease in bone mass density (BMD). Furthermore, in a recent report, Sverrisdottir compared changes in BMD from patients treated with goserelin alone versus goserelin plus tamoxifen versus tamoxifen alone versus no hormonal treatment [34Go]. During the first 2 years a decrease in BMD was observed in the three arms with hormonal treatment. It was more important in the goserelin arm, and addition to tamoxifen reduced BMD. Tamoxifen, surprisingly, reduced BMD as well, but to a lower extent.

In conclusion, early chemotherapy-induced amenorrhea is likely to have an influence on DFS in premenopausal women with endocrine-responsive breast cancer. The most important predictors of CIA are age and the chemotherapy regimen. Older women are at greater risk of CIA. With CMF and FEC regimen, the CIA incidence is between 50% and 70%. The hormonal effect of chemotherapy contributes to the cytotoxic mechanism of action. Studies on the adverse effects of these treatments should help to determine the optimal adjuvant treatment.

Received for publication December 15, 2004. Revision received March 2, 2005. Accepted for publication March 3, 2005.


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Early Breast Cancer Trialists' Collaborative Group. Polychemotherapy for early breast cancer: an overview of the randomised trials. Lancet 1998; 352: 930–942.[CrossRef][ISI][Medline]

2. Early Breast Cancer Trialists' Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials. Lancet 1998; 351: 1451–1467.[CrossRef][ISI][Medline]

3. Bines J, Oleske DM, Cobleigh MA. Ovarian function in premenopausal women treated with adjuvant chemotherapy for breast cancer. J Clin Oncol 1996; 14: 1718–1729.[Abstract/Free Full Text]

4. Poikonen P, Saarto T, Elomaa I et al. Prognostic effect of amenorrhoea and elevated serum gonadotropin levels induced by adjuvant chemotherapy in premenopausal node-positive breast cancer patients. Eur J Cancer 2000; 36: 43–48.[CrossRef][ISI][Medline]

5. Minton SE, Munster PN. Chemotherapy-induced amenorrhea and fertility in women undergoing adjuvant treatment for breast cancer. Cancer Cont 2002; 9: 466–472.

6. Pagani O, O'Neill A, Castliogne M, Gelber RD. Prognostic impact of amenorrhoea after adjuvant chemotherapy in pre-menopausal breast cancer with axillary node involvement: result of the International Breast Cancer Study Group (IBCSG) trial VI. Eur J Cancer 1998; 34: 632–634.[CrossRef][ISI][Medline]

7. Powles TJ. Prognostic impact of amenorrhoea after adjuvant chemotherapy. Eur J Cancer 1998; 34: 603–605.[CrossRef][ISI][Medline]

8. Goodwin PJ, Ennis ME, Pritchard KI et al. Risk of menopause during the first year after breast cancer diagnosis. J Clin Oncol 1999; 7: 2365–2370.

9. Padmanabhan N, Howell A, Rubens RD. Mechanism of action of adjuvant chemotherapy in early breast cancer. Lancet 1986; 23: 411–414.

10. Richards MA, O'Reilly SM, Howell A et al. Adjuvant cyclophosphamide, methotrexate, and fluorouracil in patients with axillary node-positive breast cancer: an update of the Guy's/Manchester trial. J Clin Oncol 1990; 8: 2032–2039.[Abstract/Free Full Text]

11. Fisher B, Sherman B, Rockette H et al. 1-phenylalanine mustard (L-PAM) in the management of premenopausal patients with primary breast cancer: lack of association of disease-free survival with depression of ovarian function. National Surgical Adjuvant Project for Breast and Boxel Cancers. Cancer 1979; 44: 847–857.[ISI][Medline]

12. Bonadonna G, Valagussa P, Moliterni A et al. Adjuvant cyclophosphamide, methotrexate, and fluorouracil in node-positive breast cancer: the results of 20 years of follow-up. N Engl J Med 1995; 6: 901–906.

13. Brincker H, Rose C, Rank F et al. Evidence of a castration-mediated effect of adjuvant cytotoxic chemotherapy in premenopausal breast cancer. J Clin Oncol 1987; 5: 1771–1778.[Abstract/Free Full Text]

14. Beex LV, Mackenzie MA, Raemaekers JM et al. Adjuvant chemotherapy in premenopausal patients with primary breast cancer; relation to drug-induced amenorrhoea, age and the progesterone receptor status of the tumour. Eur J Cancer Clin Oncol 1988; 24: 719–721.[CrossRef][ISI][Medline]

15. Tormey DC, Gray R, Gilchrist K et al. Adjuvant chemohormonal therapy with cyclophosphamide, methotrexate, 5-fluorouracil, and prednisone (CMFp) or CMFp plus tamoxifen compared with CMF for premenopausal breast cancer patients: an Eastern Cooperative Oncology Group Trial. Cancer 1990; 65: 200–206.[ISI][Medline]

16. Reyno LM, Levine MN, Skingley P et al. Chemotherapy induced amenorrhoea in a randomised trial of adjuvant chemotherapy duration in breast cancer. Eur J Cancer 1992; 29A: 21–23.[CrossRef][ISI][Medline]

17. Prolonged disease-free survival after one course of perioperative adjuvant chemotherapy for node-negative breast cancer. The Ludwig Breast Cancer Study Group. N Engl J Med 1989; 320: 491–496.[Abstract]

18. Del Mastro L, Venturini M, Sertoli MR, Rosso R. Amenorrhea induced by adjuvant chemotherapy in early breast cancer patients: prognostic role and clinical implications. Breast Cancer Res Treat 1997; 43: 183–190.[CrossRef][ISI][Medline]

19. A randomized trial of adjuvant combination chemotherapy with or without prednisone in premenopausal breast cancer patients with metastases in one to three axillary lymph nodes. Ludwig Breast Cancer Study Group. Cancer Res 1985; 45: 4454–4459.[Abstract]

20. Toma S, Repetto L, Giacchero A et al. Chemotherapy-induced amenorrhea and other clinical and pathological parameters in the prognosis of breast cancer patients. J Chemother 1992; 4: 321–325.[ISI][Medline]

21. Borde F, Chapelle-Marcillac I, Fumoleau P et al. Role of chemo-induced amenorrhea in premenopausal, node-positive, operable breast cancer patients: 9-year follow-up results of French Adjuvant Study Group (FASG) data base. San Antonio Meeting 2003; Poster Session: Treatment: Adjuvant Therapy, No. 138.

22. Bianco AR, Del Mastro L, Gallo C et al. Prognostic role of amenorrhea induced by adjuvant chemotherapy in premenopausal patients with early breast cancer. Br J Cancer 1991; 63: 799–803.[ISI][Medline]

23. Nabholtz JM, Pienkowski T, Mackey J et al. Phase III trial comparing TAC (docetaxel, doxorubicin, cyclophosphamide) with FAC (5-fluorouracil, doxorubicin, cyclophosphamide) in the adjuvant treatment of node positive breast cancer patients: interim analysis of the BCIRG 001 study. Proc Ann Meet Am Soc Clin Oncol 2002; 141 (Abstr).

24. Di Cosimo S, Alimonti A, Ferretti G, Cognetti F. Incidence of chemotherapy-induced amenorrhea depending on the timing of treatment by menstrual cycle phase in women with early breast cancer. Ann Oncol 2004; 15: 1065–1071.[Abstract/Free Full Text]

25. Early Breast Cancer Trialists' Collaborative Group. Tamoxifen for early breast cancer and ovarian ablation for early breast cancer (Cochrane Review). The Cochrane Library, Issue 4. Chichester, UK: John Wiley & Sons, Ltd 2004.

26. Early Breast Cancer Trialists' Collaborative Group. Ovarian ablation in early breast cancer: overview of the randomised trials. Lancet 1996; 348: 1189–1196.[CrossRef][ISI][Medline]

27. Davidson NE, O'Neill A, Vakov A et al. Chemohormonal therapy in premenopausal node-positive, receptor-positive breast cancer: An Eastern Cooperative Oncology Group phase III intergroup trial (E5188, INT-0101). Proc ASCO 2003; 15 (Abstr).

28. Pritchard KT. Adjuvant therapy for premenopausal women with breast cancer: is it time for another paradigm shift? J Clin Oncol 2002; 20: 4611–4614.[Free Full Text]

29. Jonat W, Kaufmann M, Sauerbrei W et al. Goserelin versus cyclophosphamide, methotrexate, and fluorouracil as adjuvant therapy in premenopausal patients with node-positive breast cancer: The Zoladex Early Breast Cancer Research Association Study. J Clin Oncol 2002; 20: 4628–4635.[Abstract/Free Full Text]

30. Kaufmann M, Jonat W, Blamey R et al. for the Zoladex Early Breast Cancer Research Association (ZEBRA) Trialists' Group. Survival analyses from the ZEBRA study. Goserelin (Zoladex) versus CMF in premenopausal women with node-positive breast cancer. Eur J Cancer 2003; 39: 1711–1717.[CrossRef][ISI][Medline]

31. Roche HH, Kerbrat P, Bonneterre J et al. Complete hormonal blockade versus chemotherapy in premenopausal early-stage breast cancer patients (Pts) with positive hormone-receptor (HR+) and 1 to 3 nodepositive (N+) tumor: Results of the FASG 06 trial. Proc Am Soc Clin Oncol 2000; 19: 72a (Abstr).

32. de Haes J, Olschewski M, Kaufmann et al. Quality of life in goserelin-treated versus CMF-treated pre-/perimenopausal patients with node-positive early breast cancer. J Clin Oncol 2003; 21: 4510–4516.[Abstract/Free Full Text]

33. Shapiro CL, 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.[Abstract/Free Full Text]

34. Sverrisdottir A, Fornander T, Jacobson H et al. Bone mineral density among premenopausal women with early breast cancer in a randomized trial of adjuvant endocrine therapy. J Clin Oncol 2004; 22: 3694–3699.[Abstract/Free Full Text]