The aromatase inhibitor anastrozole maintained a preventive effect for postmenopausal women at high risk for breast cancer nearly 12 years after discontinuing treatment.
Jack Cuzick, PhD
The aromatase inhibitor (AI) anastrozole (Arimidex) maintained a preventive effect for postmenopausal women at high risk for breast cancer nearly 12 years after discontinuing treatment. In data presented at the 2019 San Antonio Breast Cancer Symposium (SABCS), women assigned to anastrozole were 49% less likely to have developed breast cancer compared with women assigned to placebo.1
Jack Cuzick, PhD, director of the Wolfson Institute of Preventive Medicine in London and head of Centre for Cancer Prevention and the John Snow Professor of Epidemiology at Queen Mary, University of London said the results show that 5 years of therapy anastrozole can safely and effectively prevent breast cancer in this patient population.
“These data provide further support for use of anastrozole for breast cancer prevention in high risk postmenopausal women,” he said. “Five years of treatment—remember, treatment stops at 5 years—carries on producing continuing benefits right on to 12 years.”
Cuzick first presented results from IBIS-II at the 2013 SABCS meeting. At 5 years, investigators observed a 61% reduction in breast cancer incidence for the anastrozole group at 5 years (4.6% vs 1.8%; HR, 0.39; 95% CI, 0.27-0.58). The number needed to treat was 36, which Cuzick called “pretty good.”2
He said the preventive effect was not as great for years 5 to 12, but the difference was not significant. Anastrozole was associated with a 36% reduction in new cancers (4.4% vs 3.5%; HR, 0.64; 95% CI, 0.45-0.91; P = .014). Incidence was 5.3% in the anastrozole group compared with 8.8% in the placebo group (HR, 0.51; 95% CI, 0.39-0.66; P <.0001).
In comparison, the annual reduction rate for tamoxifen is about 30%, though Cuzick said that drug has been followed out to 20 years.
“If you put all that together, you end up with a substantial benefit over 12 years; 49%, almost half, the breast cancers were prevented,” he said. “The number needed to treat now drops to 29.”
He added that 49 patients needed to be treated with tamoxifen to prevent 1 cancer at that same period.
From 2003 to 2012, investigators recruited 3864 postmenopausal women at increased risk for developing breast cancer into the IBIS-II Prevention trial. Women with ≥2 blood relatives with breast cancer, a mother or sister who developed breast cancer before the age of 50, or a mother or sister who developed cancer in both breasts were considered high risk. Roughly half the participants (n = 1920) were randomly assigned to anastrozole for 5 years while 1944 were assigned to placebo.
Mortality data is immature with 69 deaths with anastrozole versus 70 with placebo. Investigators have not seen an improvement in overall survival, but Cuzick said “it is way too early” to draw any conclusions.
Investigators were surprised to see a dramatic reduction in the numbers of skin cancers for anastrozole (52 vs 85; OR, 0.61; 95% CI, 0.42-0.88). Cuzick said it's unclear why the AI had that effect.
Adherence was high in both groups. In the anastrozole arm, 74.6% of patients underwent the full 5 years of treatment compared with 77.0% in the placebo arm. Cuzick attributed the difference to patients dropping out of the anastrozole arm because of adverse events.
“We reported in the 5-year period there a lot of reported side effects, particularly musculoskeletal events,” he said. Most of these effects being reported are not treatment related. There's a very high rate of musculoskeletal arthralgias even in the placebo arm, indicating that we have to be very careful to interpret this.”
Cuzick presented results here in 2015 showing that recurrence was similar for postmenopausal women with ductal carcinoma in situ (DCIS) assigned to anastrozole or tamoxifen following local excision. At the time, women assigned to anastrozole had an 11% lower rate of recurrence of DCIS or invasive cancer than those randomized tamoxifen (6.6% vs. 7.4%), but the difference was not significant (P = .49).3