Phase III Study Does Not Support Use of Extended Anastrozole After Adjuvant Tamoxifen

Using extended adjuvant aromatase inhibitor (AI) therapy after 5 years of sequential endocrine therapy in all postmenopausal women with hormone receptor-positive early breast cancer is not supported by the results of a phase III randomized study.

Vivianne Tjan-Heijnen, MD, PhD

Using extended adjuvant aromatase inhibitor (AI) therapy after 5 years of sequential endocrine therapy in all postmenopausal women with hormone receptor-positive early breast cancer is not supported by the results of a phase III randomized study. First results from the DATA study show no advantage to extended anastrozole on the primary endpoint of 5-year adapted disease-free survival (DFS), reported Vivianne Tjan-Heijnen, MD, PhD, at the 2016 San Antonio Breast Cancer Symposium.

In a subgroup analysis that was not prespecified, extended anastrozole did show a significant effect on adapted DFS in women who were estrogen receptor (ER)-positive and progesterone receptor (PR)-positive,HER2-negative, node-negative, and who received neoadjuvant chemotherapy. Adapted DFS was defined as the DFS starting as of 3 years after randomization. “This was an unplanned analysis, but I think this is very useful for daily clinical practice,” said Tjan-Heijnen, professor in the division of medical oncology in the department of Internal Medicine at Maastricht University Medical Center, Maastricht, The Netherlands.

In the overall analysis, 5-year adapted DFS was 83.1% in the group randomized to 6 years of anastrozole, compared with 79.4% in those randomized to only 3 years of anastrozole after 2 to 3 years of adjuvant tamoxifen, for a hazard ratio of 0.79, which was not significant (P= .07).

In DATA, 1,912 women with M0 breast cancer who were postmenopausal and ER- and/or PR-positive with no sign of disease recurrence and who received 2 to 3 years of adjuvant tamoxifen were randomized to 3 years (n = 833) or 6 years (n = 827) of 1 mg anastrozole per day. Minimum follow-up was at least 6 years after randomization. Patients were stratified by nodal status, RR/PR status,HER2status, and duration of tamoxifen.

“It’s very important to know that for the first 3 years after randomization, all patients received the same treatment, so we decided to have the adaptive DFS as the primary endpoint…which [means] that we account for events starting beyond a year after randomization,” said Tjan-Heijnen.

The adapted DFS endpoint events included breast cancer recurrence (local, regional, or distant); a second primary cancer, including contralateral breast cancer and cancers other than basal-cell or squamous-cell carcinoma of the skin, or carcinoma in situ of the cervix; and death from any cause.

Most patients had larger tumors or had nodal positive disease. About 54% in each arm had pathologic stage II tumors, pathologic stage III/IV tumors, and about two thirds in each arm had pathologic nodal status of stage I or stage II/III. Three fourths of the patients had both ER-positive and ER-positive disease, and 90% hadHER2-negative disease. “So it seems to be a hormone-sensitive population,” Tjan-Heijnen said.

Although the 5-year adapted DFS of 83.1% in the 6-year anastrozole group, and 79.4% in the 3-year group did not achieve significance, there were several subgroups that did show significant benefit to extended anastrozole. The HR in favor of 6 years of anastrozole for patients with larger (pathologic tumor stage II or pathologic tumor stage III/IV) tumors was 0.72 (95% CI, 0.51- 0.98) and the difference in adapted 5-year DFS was about 6%, “which I think is also clinically meaningful,” she said.

The same advantage to 6 years of anastrozole was observed in the node-positive group (HR 0.72; 95% CI, 0.52-1.00), patients with both ER- and PR-positive disease (HR 0.70; 95% CI, 0.53-0.92), patients with HER2-negative status (HR 0.79; 95% CI, 0.61-1.03), and those who received neoadjuvant chemotherapy (HR 0.68; 85% CI, 0.49-0.92).

The 5-year adapted DFS for patients who were ER-positive, PR-positive, HER2-negative, node-negative and who received neoadjuvant chemotherapy was 86.0% in those randomized to 6 years compared to 75.9% in those randomized to 3 years of anastrozole (HR 0.58;P= .01).

There was no difference in 5-year adapted overall survival: 90.8% compared to 90.4% in the 6-year and 3-year groups, respectively, but the median adapted follow-up was only 4.1 years at the time of presentation.

The 6-year anastrozole group had a slightly higher rate of predefined adverse events compared to the 3-year group, such as arthralgia/myalgia (57.6% vs 51.9%), bone fracture (9.8% vs 7.4%), osteopenia/osteoporosis (20.9% vs 16.5%), and cardiovascular events including arrhythmia (13.4% vs 12.9%).

About 80% of patients in the 3-year arm were able to complete treatment, compared with 60% in the 6-year arm. Occurrence of adverse events was the reason for early termination in 23% of the 6-year am and 14.4% in the 3-year arm.

A follow-up analysis will be performed when all patients have reached a minimum adapted follow-up of 9 years.


Tjan-Heijnen VC, Van Hellemond IE, Peer PG, et al. First results from the multicenter phase III DATA study comparing 3 versus 6 years of anastrozole after 2-3 years of tamoxifen in postmenopausal women with hormone receptor-positive early breast cancer. Presented at: 2016 San Antonio Breast Cancer Symposium; December 6-10; San Antonio, TX. Abstract S1-03.

The rationale for DATA at the time of its design in 2006 is the elevated risk of breast cancer recurrence (2% to 4% per year for up to 15 years), despite the use of 5 years of endocrine therapy. Extended use of an AI after 5 years of adjuvant tamoxifen led to an improvement in DFS in the MA17, NSABP-833, and ABCSG6a clinical trials, but no studies have examined the use of extended AI in patents with 5 years of sequential endocrine therapy (tamoxifen followed by an AI).