Anastrozole vs Tamoxifen to Prevent DCIS Recurrence: Key Difference May Be Side-Effect Profile

Get Permission


  • Anastrozole and tamoxifen lead to similar rates of recurrence in older women with surgically excised DCIS.
  • These hormonal agents have different side-effect profiles, which should be considered in shared decision-making when selecting treatment.

If your patient were an older woman opting for hormonal therapy after surgically excised ductal carcinoma in situ (DCIS), which hormonal agent would you recommend for the prevention of disease recurrence: tamoxifen or anastrozole? According to long-term follow-up of the large, randomized IBIS II-DCIS trial presented at the 2020 San Antonio Breast Cancer Symposium, both drugs achieved similarly low rates of recurrence in the active treatment phase and posttreatment phase.1 Thus, the choice between those two options may come down to side-effect profile as it relates to individual patient preference.

For the full follow-up period (median = 11.6 years), 103 recurrences were reported in the anastrozole arm and 118 in the tamoxifen arm. The numerical trend favored anastrozole, but the difference between the two treatment arms was not statistically significant.

As expected, more endometrial and ovarian cancers were reported in the tamoxifen arm: there were 2 endometrial cancers in the anastrozole arm and 12 in the tamoxifen arm, and 1 vs 8 ovarian cancers. Patients on 5 years of anastrozole experienced a significantly higher rate of fractures, strokes, and transient ischemic events compared with those on tamoxifen therapy. More reports of pulmonary embolism, deep-vein thrombosis, and myocardial infarction occurred in the tamoxifen arm compared with anastrozole, but these differences were not statistically significant.

“Although our analysis did not show any difference in terms of recurrences between anastrozole and tamoxifen, it shows that an improved understanding of adverse-event profiles will help patients with estrogen receptor–positive DCIS make an informed decision about treatment,” stated lead author Ivana Sestak, MD, of the Wolfson Institute of Preventive Medicine, Queen Mary University, London.

“[Our analysis] shows that an improved understanding of adverse-event profiles will help patients with estrogen receptor–positive DCIS make an informed decision about treatment.”
— Ivana Sestak, MD

Tweet this quote

In this analysis, Dr. Sestak did not discuss patient-reported side effects, which have been well described with both hormonal agents. In particular, anastrozole commonly leads to treatment discontinuation due to muscle aches.

Dr. Sestak explained that large trials in hormone receptor–positive invasive breast cancer—for example, the ATAC trial 2—have shown that anastrozole reduces the risk of recurrence to a greater extent than tamoxifen. She and her co-investigators wanted to compare these two treatments at an earlier stage of breast cancer, DCIS, which was once called stage 0.

The study Dr. Sestak presented updated the blinded efficacy of anastrozole vs tamoxifen in preventing breast cancer recurrences, with an emphasis on the posttreatment phase. The authors wanted to explore the efficacy of these two hormonal agents on breast cancer subtypes and compare adverse events occurring during and after 5 years of treatment.

Study Details

Between March 2003 and February 2012, the investigators recruited 2,980 postmenopausal women aged 40 to 70 years who had undergone local excision for estrogen receptor–positive DCIS within the past 6 months. Women with atypical hyperplasia and lobular carcinoma in situ were also enrolled in the trial and accounted for less than 2% of the study population. Participants were randomly assigned 1:1 to receive anastrozole at 1 mg/d (n = 1,471) vs tamoxifen at 20 mg/d (n = 1,509) for 5 years.

The median follow-up for the active treatment and posttreatment analyses was 11.6 years. Women were followed during the posttreatment phase with clinic visits, questionnaires, and registry data.

At baseline, demographic and disease characteristics were well balanced between the two treatment arms. The median age was 60.4 years; the median body mass index was 26.5 kg/m2; and about 45% had ever used hormone replacement therapy. Breast cancer grades and mean tumor size were similar in the two arms. The majority of patients in both arms had radiotherapy as their initial treatment, and about 28% had a hysterectomy.

Both invasive and DCIS recurrences were included in the primary endpoint of the trial. At 5 years of follow-up during the active phase of treatment, 37 recurrences were reported with anastrozole compared with 45 for tamoxifen, for a nonsignificant relative risk reduction of 16% and an absolute risk reduction of 0.4%.

For the full follow-up period at a median follow-up of 11.6 years, a nonsignificant 11% reduction in breast cancer recurrences was observed favoring anastrozole, with an absolute difference of 1.2% between treatment arms.

Active Treatment Period vs Posttreatment Period

“For all disease recurrences, we observed nonsignificant reductions with anastrozole in both follow-up periods. When we looked at estrogen receptor–positive recurrences, we observed a significant 44% reduction [during treatment] with anastrozole vs tamoxifen,” Dr. Sestak stated.

The number of estrogen receptor–positive recurrences for the active treatment period was 18 with anastrozole vs 33 with tamoxifen (hazard ratio [HR] = 0.56, 95% confidence interval [CI] = 0.31–0.99), and for the follow-up years, the number of events was 40 vs 49 (HR = 0.83, 95% CI = 0.55–1.27). There was no significant difference in estrogen receptor–negative recurrences.

There were 9 HER2-negative recurrences in the anastrozole arm and 23 in the tamoxifen arm during the active treatment period, and 37 and 39, respectively, during the posttreatment phase. Similar differences were observed for invasive and DCIS recurrences in both follow-up periods.

Looking at invasive and DCIS recurrences in different breast cancer subtypes, there was no significant effect of anastrozole vs tamoxifen on reducing invasive recurrences: 66 in the anastrozole arm and 76 in the tamoxifen arm.

“There was an indication that patients with estrogen receptor–positive and HER2-negative disease and tumors smaller than 10 mm did better on anastrozole,” she said.

Adverse-Event Profiles

Other cancers were found in 7% of the anastrozole arm and 7.5% of the tamoxifen arm. The majority of endometrial and ovarian second cancers were reported during the active treatment phase. For anastrozole recipients, they were found only during active treatment, and for tamoxifen, they occurred during active treatment and after treatment.

There were 2 cases of endometrial cancer with anastrozole and 12 with tamoxifen (P = .0086). There was one ovarian cancer in the anastrozole group and three in the tamoxifen group (P = .022). There were a total of 5 gynecologic cancers with anastrozole and 21 with tamoxifen (P = .0021). A nonsignificant increase in gastrointestinal cancers and lung cancers was observed in the anastrozole arm.

A 34% increase in fractures was observed in patients treated with anastrozole (P = .013). Strokes were reported in 15 vs 5 patients (P = .021), and transient ischemic attacks occurred in 15 and 5 patients, respectively (P = .021). A nonsignificant numerical difference favoring anastrozole over tamoxifen was observed for pulmonary embolism, deep-vein thrombosis, and myocardial infarction.

There was no significant difference in the total number of deaths between the two treatment arms (61 vs 66; HR = 0.94, 95% CI = 0.67–1.33; P = .74). Three patients in each arm died of breast cancer. “Other cancer” was the cause of death in 23 and 34 patients, respectively.

“The late positive effect of anastrozole reported in the NSABP B-35 trial was not seen in IBIS II-DCIS. We could not replicate the reduction in risk of recurrence with anastrozole vs tamoxifen in younger vs older patients.” Dr. Sestak stated. “We did, however, see clear differences in adverse events between anastrozole and tamoxifen.

For this updated analysis, no further data on adverse events were collected beyond the active treatment period. In e-mail correspondence with The ASCO Post, Dr. Sestak responded to a specific question about the relationship of adherence to the side-effect profiles of anastrozole vs tamoxifen.

“There was an indication that patients with estrogen receptor–positive and HER2-negative disease and tumors smaller than 10 mm did better on anastrozole.”
— Ivana Sestak, MD

Tweet this quote

“We reported detailed side effects in our previous publication, which focused on the active treatment period (5 years).3 We observed significant increases in musculoskeletal events (predefined side effect), such as arthralgia and joint pain/stiffness, in women on anastrozole compared with tamoxifen (57% vs 49%, P < .0001). These side effects are common in women on aromatase inhibitors and can be bothersome to women on these drugs. We also know from our study [and many other preventive and adjuvant studies] that these events lead to greater nonadherence, which is a problem, since for best effectiveness in terms of breast cancer [recurrence] reduction, women should stay on the drug,” Dr. Sestak explained.

“I agree that side effects are a big issue for postmenopausal women on an aromatase inhibitor, and future research should be directed toward supporting decision-making at the point of uptake, as well as ensuring that mechanisms are in place to promote persistence among women who have initiated therapy,” Dr. Sestak wrote in her e-mail. 

DISCLOSURE: Dr. Sestak reported no conflicts of interest.


1. Sestak I, Cuzick J, Bonanni L, et al: Updated long-term results of anastrozole versus tamoxifen for the prevention of breast cancer in postmenopausal women with locally excised ductal carcinoma-in-situ. 2020 San Antonio Breast Cancer Symposium. Abstract GS2-02. Presented December 9, 2020.

2. Cuzick J, Sestak I, Baum M, et al: Effect of anastrozole and tamoxifen as adjuvant treatment for early-stage breast cancer: 10-year analysis of the ATAC trial. Lancet Oncol 11:1135-1141, 2010.

3. Forbes JF, Sestak I, Howell A, et al: Anastrozole versus tamoxifen for the prevention of locoregional and contralateral breast cancer in postmenopausal women with locally excised ductal carcinoma in situ (IBIS II-DCIS): A double-blind, randomized controlled trial. Lancet 387:866-873, 2016.


Related Articles

Expert Point of View: Hatem Soliman, MD

Hatem Soliman, MD

Hatem Soliman, MD

“The updated long-term follow-up from IBIS II showed that tamoxifen and anastrozole have similar efficacy in preventing breast cancer recurrences in patients diagnosed with ductal carcinoma in situ. However, there were differences in side effects consistent with the known...