Study Finds No Difference in Disease Recurrence Between Tamoxifen, Anastrozole in Postmenopausal Women With DCIS


An analysis of tamoxifen versus anastrozole showed no significant difference in terms of disease recurrence in postmenopausal women with locally excised ductal carcinoma in situ; however, the findings may highlight the understanding of associated toxicities.

Tamoxifen, compared with anastrozole, did not induce a significant difference in disease recurrence in postmenopausal women with locally excised ductal carcinoma in situ (DCIS); according to data from a study presented during the 2020 San Antonio Breast Cancer Symposium (SABCS).

However, the findings highlights oncologists’ understanding of the agents’ associated toxicities, which can inform care decisions for these patients.

“Although our analysis did not show any significant difference in terms of recurrences between anastrozole and tamoxifen, it shows that an improved understanding of AE [adverse events] profiles will help patients with HR-positive DCIS to make an informed decision regarding their treatment,” said lead investigator Ivana Sestak, PhD, Queen Mary University of London, during her presentation at SABCS.

The purpose of this study was to update the blinded impact of these 2 treatments on breast cancer recurrences, and there was a special focus on the post-treatment period. The study explored the efficacy between tamoxifen versus anastrozole across several breast cancer subtypes and compared the AEs between the 2 therapies.

Overall, 2980 postmenopausal patients aged 40 to 70 years with locally excised estrogen receptor (ER)-positive DCIS and atypical hyperplasia/LCIS (lobular carcinoma in situ), who were randomized to receive either anastrozole 1 mg/day plus tamoxifen placebo (n = 1471) or tamoxifen 20 mg/day n = 1509) plus anastrozole placebo for 5 years. The primary end point was breast cancer recurrence, both invasive and DCIS, and secondary end points included ER-positive breast cancer recurrence, breast cancer laterality, breast cancer and all-cause mortality, other cancers, and cardiovascular/thromboembolic events, and fractures.

The cut-off date for the analysis was September 30, 2020, and the median follow-up for the study was 11.6 years.

The median age of patients was 60.4 years in the anastrozole arm (interquartile range [IQR], 56.4-64.5) and 60.3 in the tamoxifen arm (IQR, 55.8-64.5), and the median body mass index was 26.7 kg/m2 (IQR, 23.5-30.4) and 26.6 kg/m2 (IQR, 23.7-30.2), respectively. The mean tumor size was 17.5 mm in the anastrozole arm and 17.6 mm in the tamoxifen arm. Overall, 20.2% and 18.7% in the anastrozole and tamoxifen arms were low grade, 41.8% and 41.5% were intermediate, and 37.4% and 39.4% were high, respectively.

Approximately 46.8% of patients in the anastrozole and 44.2% of patients in the tamoxifen arm had received prior HRT before study entry. The majority of patients in both arms had received radiotherapy as their initial treatment, and 28.0% and 27.4% had a hysterectomy, respectively.

After 5 years of follow-up, 37 breast cancer recurrences occurred in the anastrozole arm and 45 in the tamoxifen arm (HR, 0.84; 95% CI, 0.54-1.30; P =.437), and the rate of recurrence at 5-years was 2.6% for anastrozole (95% CI, 1.9-3.5) versus 3.0% with tamoxifen (95% CI, 2.3-4.1), which translated to a 0.4 percentage point absolute risk reduction associated with anastrozole.

For the full follow-up period, 103 recurrences were observed in the anastrozole arm and 118 in the tamoxifen arm (HR, 0.89; 95% CI, 0.69-1.16; P =.437). The recurrence rate with anastrozole versus tamoxifen, respectively, was 8.5% (95% CI, 6.9-10.5) and 9.7% (95% CI, 7.9-11.8), which translated to an 11% reduction and a 1.2% absolute risk reduction with anastrozole. However, the absolute risk reductions in both follow-up periods were not significant.

“We then went on to look at the active treatment period versus the post-treatment periods separately,” said Sestak. “For all recurrences, we observed nonsignificant reductions with anastrozole in both follow-up periods, and when we looked at ER-positive recurrences, we observed a significant 44% reduction with anastrozole compared to tamoxifen.”

The number of recurrence events for 0 to 5 years during the active treatment period in the anastrozole versus tamoxifen arms was 37 versus 45, respectively (HR, 0.84; 95% CI, 0.54-1.30), and for 5+ years following active treatment, the number of events was 66 versus 73 (HR, 0.92; 95% CI, 0.66-1.29). In terms of ER-positive recurrences, the number of events for the active period was 18 with anastrozole versus 33 with tamoxifen (HR, 0.56; 95% CI, 0.31-0.99), and for the follow-up years, the number of events was 40 versus 49 (HR, 0.83; 95% CI, 0.55-1.27). For ER-negative recurrences, the increase with anastrozole after the active follow-up period was not significant.

In terms of HER2-negative events, 9 occurred in the anastrozole arm and 23 in the tamoxifen arm during the 5 active years of follow-up (HR, 0.40; 95% CI, 0.19-0.87), whereas 37 and 39 occurred during the longer follow-up post-treatment (HR, 0.97; 95% CI,0.62-1.52). Similar nonsignificant differences were observed for invasive and DCIS recurrences in both follow-up periods.

Invasive and DCIS recurrences were analyzed in different breast cancer subtypes. Overall, there was no significant effect of anastrozole on reducing invasive recurrences, with 66 events occurring in the anastrozole arm versus 76 in the tamoxifen arm (HR, 0.89; 95% CI, 0.64-1.24; P =.048).

Despite lacking a significant difference compared with tamoxifen, anastrozole more effectively reduced invasive recurrence in ER-positive and HER2-negative breast cancer subgroups, as well as those with tumors smaller than 10 mm in size. No differences were observed between the 2 treatments for DCIS recurrences across breast cancer subgroups. However, there was an indication in the findings that the ER-positive subgroup and those with tumors larger than 20 mm were more effectively reduced with anastrozole.

Other cancers were observed in 102 patients (7.0%) in the anastrozole arm and 112 (7.5%) of the tamoxifen arm (odds ratio [OR], 0.93; 95% CI, 0.70-1.24; P =.61). In particular, 2 in the anastrozole arm versus 12 patients in the tamoxifen arm had endometrial cancer (OR, 0.17; 95% CI, 0.02-0.77; P =.0086), and 1 versus 8, respectively, had ovarian cancer (OR, 0.13; 95% CI, 0.03-0.96; P =.022). In the anastrozole arm, these events only occurred after the active follow-up period. A nonsignificant increase was also noted in gastrointestinal cancers and lung cancers in the anastrozole arm compared with tamoxifen.

In terms of AEs, fractures were reported in 183 patients in the anastrozole arm versus 145 in the tamoxifen arm (OR, 1.34; 95% CI, 1.06-1.70; P =.013), strokes were reported in 15 versus 5 patients (OR, 3.10; 95% CI< 1.07-10.92; P =.021), and transient ischemic attacks were reported in 15 versus 5 (OR, 3.10; 95% CI< 1.07-10.92; P =.021). Pulmonary embolism, deep vein thrombosis, and myocardial infarction occurred more frequently in the tamoxifen arm, but these differences were nonsignificant compared with anastrozole.

Overall, 61 deaths (4.2%) with anastrozole and 67 (4.5%) with tamoxifen were reported (OR, 0.93; 95% CI, 0.64-1.35). The cause of death was breast cancer in 3 patients for each treatment arm. Other cancer was the of death in 23 patients in the anastrozole arm versus 34 in the tamoxifen arm (OR, 0.66; 0.37-1.17), cardiac/stroke in 7 versus 8, and other/unknown in 28 versus 21, respectively. Endometrial cancer was the cause of death in 4 patients in the tamoxifen arm and ovarian cancer in 2, although neither were causes in the anastrozole arm.

In conclusion, the investigators did not observe any effects on breast cancer deaths with anastrozole. However, there are clear differences in AEs between the 2 therapies; with tamoxifen, an excess of endometrial and ovarian cancers was observed, whereas anastrozole induced an excess of fractures and strokes.


Sestak I, Cuzick J, Bonanni B, et al. 12 year results of anastrozole versus tamoxifen for the prevention of breast cancer in postmenopausal women with locally excised ductal carcinoma. Presented at: 2020 San Antonio Breast Cancer Symposium; December 8, 2020; Virtual. Abstract GS2-02.

Related Videos
The August CancerNetwork Snap Recap takes a look back at key FDA news updates, as well as expert perspectives on the chemotherapy shortage.
Common adverse effects following treatment with lenvatinib plus pembrolizumab in the phase 3 CLEAR study include diarrhea, hypertension, and fatigue, according to Thomas E. Hutson, DO, PharmD, FACP.
Ann H. Partridge, MD, MPH, talks about how fertility preservation can positively impact the psychosocial health in patients with breast cancer.
Lenvatinib in combination with pembrolizumab appears to raise no new safety signals in patients with advanced clear cell renal cell carcinoma after 4 years of follow-up in the phase 3 CLEAR study.
According to Thomas E. Hutson, DO, PharmD, FACP, 4-year follow-up data from the phase 3 CLEAR study confirm the maintained benefits of lenvatinib plus pembrolizumab in patients with advanced renal cell carcinoma.
Findings from the phase 3 MIRASOL trial support mirvetuximab soravtansine as a standard treatment option for platinum-resistant ovarian cancer, according to Ritu Salani, MD.
Rana R. McKay, MD discusses presentations of interest that were presented at the 2023 Kidney Cancer Research Summit, including a discussion on how PET imaging may identify which patients with renal cell carcinoma may respond to immunotherapy.
A better understanding of tumor biology may be necessary for identifying novel non-immunotherapy–based therapeutic strategies for patients with renal cell carcinoma, according to Rana R. McKay, MD.
Probiotics and other agents targeting fatty acid oxidation are also under evaluation as treatment options for patients with renal cell carcinoma, according to Rana R. McKay, MD.
Other angiogenic agents are also under investigation in renal cell carcinoma, according to Rana McKay, MD, who indicates it will be interesting to see how they compare with belzutifan.
Related Content