Adjuvant ovarian suppression combined with tamoxifen or anastrazole, alone or in combination with zoledronic acid, in premenopausal women with endocrine-responsive, stage I and II breast cancer: First efficacy results from ABCSG

Reviewer: Arpi Thukral, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 1, 2008

Presenter: M. Gnant
Presenter's Affiliation: Medical University of Vienna
Type of Session: Scientific


  • In hormone receptor-positive breast cancers, growth of the tumor is stimulated by estrogens. Decreasing the production of estrogens therefore leads to suppression of recurrence, and drugs such as tamoxifen and aromatase inhibitors are useful for this purpose.
·         In contrast to pre-menopausal women whose estrogen is produced in the ovaries, in post-menopausal women most estrogen is produced in the adrenal gland from the conversion of androgens.
o       Therefore, ovarian suppression is needed for aromatase inhibitors to be effective in premenopausal women.
·         Tamoxifen had been standard treatment for localized estrogen receptor-positive breast cancer; however the ATAC trial has shown that clinical results are superior with an AI (specifically anastrazole) in postmenopausal women.  
·         Aromatase inhibitors are generally not used to treat breast cancer in premenopausal women since ovarian suppression would be necessary. No clear data is available regarding the use of AI’s in premenopausal women.   A comparison of TAM and ANA in premenopausal patients in a large randomized trial is warranted. 
o       Goserelin, a gonadotropin-releasing hormone analogue, can be used to stimulate ovarian suppression in pre-menopausal women.
  • In preclinical and early clinical studies, zoledronic acid (ZOL) is a very potent bisphosphanate, and has been shown to have a direct anti-tumor effect and anti-metastatic effect by inhibiting tumor growth, stimulating immune response and inhibiting angiogenesis.  It has also been shown to decrease the risk of osteoporosis in postmenopausal women. 
  • The Austrian Breast and Colorectal Cancer Study Group Trial 12 (ABCSG-12) was undertaken to examine the efficacy of goserelin in combination with ANA or TAM +/- ZOL in pre-menopausal women with hormone receptor-positive breast cancer.   

Materials and Methods

  • Between 1999 and 2006, 1,803 premenopausal women with Stage I or II hormone receptor-positive breast cancer were randomized to goserelin (3.6 mg q 28 days SC) + TAM (20 mg/day PO) and goserelin +ANA (1 mg/day PO) +/-ZOL (4 mg IV q 6 months). 
  • Essentially, the trial has four arms comparing goserelin/ tamoxifen to goserelin/ anastrazole, with or without zoledronic acid, and followed a 2x2 factorial design.
  • Eligibility criteria included Stage I or II breast cancer, <10 + nodes, and no chemotherapy except neoadjuvant chemotherapy. 
  • The treatment duration was 3 years, and patients were then followed for 2 additional years. 
  • The primary endpoint was DFS for both the comparisons of TAM vs. ANA and ZOL vs. no ZOL. 
  • Secondary endpoints included relapse-free survival (RFS), overall survival (OS), and safety. 
  • A Cox proportional hazard regression model was used to assess differences between the various groups. 
    • Since there were 2 pair-wise comparisons being performed, a significance level of 2.5% was used for the primary endpoint (DFS) analysis using a Bonferroni correction. 
    • A 5% significance level was still used for the secondary endpoint analysis. 


  • Patient characteristics were balanced, and there were no significant differences between the groups. 
  • The median age of patients was 45 years. 
  • About 70-75 patients had Stage I tumors, and the rest were Stage II. Only 30% of patients were node positive. 
  • At a median follow up of 60 months in March 2008, 137 (7.6%) DFS events and 42 (2.3%) deaths occurred. There was no significant difference seen between for DFS between the ANA (n=900) group and the TAM (n=903) group, with a HR of 1.1 (0.79, 1.54), p=0.59. Analysis of secondary endpoints also did not show any differences between the TAM and ANA groups. 
  • ANA was associated with fewer thrombotic events and endometrial changes compared with TAM.
  • The ZOL group had 904 patients, while the no ZOL group had 899 patients. The addition of ZOL significantly reduced the risk of DFS events by 36% (HR=0.64: 95% CI -0.46, 0.91, p=0.011) compared with endocrine therapy alone. This improvement in DFS was seen for all types of events, including bony metastases, contralateral breast cancer, and other sites of metastatic disease. 
  • There was also a statistically significant difference in RFS shown between the ZOL and no ZOL group. The addition of ZOL reduced the risk of RLS events by 35% (HR= 0.65: 95% CI 0.46,0.92; p=0.015) compared with no ZOL treatment.
  • For OS, there was not a significant difference seen as the number of deaths was fortunately small. However, a non-significant trend favoring ZOL treatment was seen for OS (HR=0.6: 95% CI -0.32, 1.11, p=0.10. 
  • Treatment of ZOL was well-tolerated and side effects were as expected. There were no significant associations with serious adverse events. 

Author's Conclusions

  • There were no significant differences seen in DFS, RFS, or OS between ANA and TAM in pre-menopausal women with Stage I or II breast cancer. This may be related to the effects of ovarian suppression with goserelin in premenopausal patients. 
  • It was also seen that few potentially life-threatening serious adverse events were noted with ANA compared to TAM. 
  • The addition of ZOL to endocrine therapy significantly prolonged DFS and RFS compared with endocrine therapy alone in pre-menopausal women with Stage I or II breast cancer.
  • The authors further conclude that this demonstrates we can utilize direct antitumor activity and immune activation in conjunction with the effects of hormonal treatment to increase DFS. 

Clinical/Scientific Implications

  • This prospective randomized trial provides relevant information regarding selection of hormonal treatment in pre-menopausal patients, and shows that the use of bisphosphonates, such as zoledronic acid, increases the effectiveness of current hormonal therapies. 
  • The strengths of this study are that it was done with a 2x2 factorial design; patients had virtually no chemotherapy background; and it was performed with an intent-to-treat analysis. However, the trial's weaknesses include that it was not double blinded; there was no information on Her2 status; and it was underpowered for the endocrine question. 
Endocrine Question:
  • Although this study showed that there were no differences between TAM and ANA in premenopausal patients, the authors noted that the analysis was severely underpowered, since enough DFS events did not occur to note differences. For this reason, a benefit should not yet be ruled out. Ongoing trials such as the SOFT and TEXT trials are also comparing TAM with AIs in treatment of breast cancer in premenopausal women, and will hopefully help us better understand this question.
  • The authors did note that at 5 years, 137 (7.6%) DFS events and 42 (2.3%) deaths occurred. This 94% DFS rate and 98.2% OS rate at 5 years for the overall trial are reassuring, but may also have been the reason why the endocrine question was unable to be answered. 
  • Other questions to consider based on this data would be whether potency of the AI matters more in pre-menopausal women, or whether goserelin + TAM is better than TAM alone. Also, the authors did not address for how long this therapy should optimally continue. Lastly, would a Tam à AI switch be better than either treatment alone in pre-menopausal women?
Bisphosphonate question:
  • The authors demonstrated a significant improvement in DFS and RFS with the addition of ZOL to endocrine therapy. The authors have done an excellent job in designing and implementing a large enough study to evaluate this question. Their findings are suggestive that ZOL does in fact have direct anti-tumor activity as seen in preclinical trials.  It was also interesting to see that the effects of ZOL were not limited to bone. It is a well-tolerated and safe drug which appears to be promising in breast cancer.
  • However, the data for these results are only based on 137 events (since patients did so well), so these conclusions are made with caution. We need to await results of the ongoing AZURE trial, which randomizes Stage II breast cancer patients (N=3349) to ZOL vs. nil for 5 years. The interim analysis has 472 DFS events, which is triple the number in this study. 
  • Further questions to consider regarding this trial include:
    • 1) how long should the bisphosphonates be given?
    • 2) what is the mechanism of the beneficial effect of ZOL?
    • 3) what are implications for other tumor types. 
  • Longer follow up is needed for OS data before adjuvant treatment with ZOL should be considered as an improvement to the standard of care for premenopausal women with breast cancer.
  • This study is not yet a practice-changing trial, but it does open the door for studying yet another strategy for early breast cancer. 


Frequently Asked Questions