SAN ANTONIOA new analysis of the ATAC (Arimidex, Tamoxifen,
Alone or in Combination) trial shows that the aromatase inhibitor anastrozole (Arimidex)
was more effective in preventing relapses than tamoxifen in receptor-positive
breast cancer patients (estrogen and/or progesterone positive [ER+ and/or
PR+]), with the greatest effect seen in ER-positive/PR-negative (ER+/PR-)
women, said Mitch Dowsett, PhD, professor of biochemical endocrinology, Royal
Marsden Hospital, London, UK, on behalf of the ATAC Trialists’ Group.
The study aim, Dr. Dowsett said at the 26th Annual San
Antonio Breast Cancer Symposium (abstract 4), was to test the hypothesis that
the relative benefit of anastrozole over tamoxifen (or the combination) in the
ATAC trial differed according to PR status.
ATAC included 9,366 postmenopausal women with invasive
breast cancer, recruited based on their suitability for hormone therapy: 84%
were hormone-receptor positive (ER+ and/or PR+) with 8% receptor negative and
8% receptor status unknown. The patients were randomized 1:1:1 after completion
of primary therapy to 5 years of tamoxifen (20 mg/d) or anastrozole (1 mg/d)
with appropriate placebos or their combination. The primary endpoint was
disease-free survival, with relapse-free survival as a secondary endpoint.
In the new analysis, Dr. Dowsett said, "we are concentrating
on the relapse-free survival data because that analysis reveals the impact of
tumor biological features on the ability of the treatments to interact with the
Prior analysis of outcomes among hormone-receptor-positive
women at 47 months had shown a highly significant (P = .007) 0.78 hazard
ratio (22% reduction in breast cancer events) for anastrozole, compared with
For this retrospective analysis, ER and PR status were known
in 7,993 women (86% of the cohort). Most patients (71%) were ER+/PR+, with 17%
ER+/PR-, 3% ER-/PR+, and 9% ER-/PR-.
Among the ER+/PR+ women, the hazard ratio for reduction in
breast cancer events for anastrozole vs tamoxifen was 0.82 (P = .091).
Outcomes for tamoxifen vs tamoxifen plus anastrozole were nearly identical. For
ER-/PR+ patients, the hazard ratio was 0.79 in favor of anastrozole. For
ER+/PR- women, however, the hazard ratio was significantly halved to 0.48 in
women receiving anastrozole, compared with tamoxifen (P < .001).
Presenting Forest plots, Dr. Dowsett pointed out that, "for
all three receptor-positive subgroups, the point estimate was to the left,
suggesting a better performance for anastrozole than for tamoxifen, with no
differences in the ER-/PR- group." When adjustments were made for baseline
nodal status, tumor size and grade, and use of adjuvant chemotherapy, the
hazard ratio for anastrozole vs tamoxifen in ER+ women was 0.80 in those who
were also PR+ and 0.48 in those who were PR-.
Noting the greater efficacy of anastrozole in ER+ women, and the halving of
relapses in ER+/PR- women, Dr. Dowsett concluded that, "these data generate the
hypothesis that aromatase inhibitors may have greater efficacy relative to
tamoxifen in patients with ER+/PR- tumors than in those with ER+/PR+ tumors."
He added that before these findings affect treatment selection, they should be
confirmed in other adjuvant trials of aromatase inhibitors.