ATAC: What the Breast Cancer Trial Found

The question, in one sentence

Most breast cancers are hormone-receptor-positive, meaning the cells carry receptors for estrogen and use it to grow. Treatment after surgery aims to cut that supply off.

Tamoxifen does it one way. It sits in the estrogen receptor and blocks it.

Anastrozole does it another. After menopause, the ovaries stop making most estrogen and an enzyme called aromatase takes over, converting other hormones into estrogen in fat and muscle. Anastrozole blocks that enzyme, so less estrogen is made at all.

ATAC — Arimidex, Tamoxifen, Alone or in Combination — asked which approach was better after surgery in postmenopausal women.

How it was built

ATAC randomly assigned 6,241 postmenopausal women with early-stage breast cancer: 3,125 to anastrozole at 1 mg daily and 3,116 to tamoxifen at 20 mg daily, both taken by mouth for five years. Both were fixed trial amounts; a tablet prescribed for you today may not match.

The amounts describe how the trial was run. Your own doctor decides which hormone tablet you take and for how long.

Of those, 5,216 had hormone-receptor-positive tumors. That subgroup matters, because these drugs have no mechanism to work without receptors, and it is where the trial's clearest answer lies.

The primary endpoint was disease-free survival. The 10-year analysis reported here covers a median follow-up of 120 months, with 24,522 woman-years in the anastrozole group and 23,950 in the tamoxifen group.

Fractures and serious adverse events kept being recorded after treatment ended, in a masked fashion.

What it found

In the full study population, anastrozole beat tamoxifen on disease-free survival: hazard ratio 0.91, 95% confidence interval 0.83 to 0.99, p = 0.04.

Time to recurrence favored anastrozole more strongly, at 0.84. Time to distant recurrence was 0.87.

In hormone-receptor-positive women the effect was larger. Disease-free survival hazard ratio was 0.86 (p = 0.003) and time to recurrence 0.79 (p = 0.0002).

The absolute numbers are the ones worth carrying. In hormone-receptor-positive women, the gap in recurrences between anastrozole and tamoxifen was 2.7 percentage points at 5 years and 4.3 percentage points at 10 years. The difference widened after treatment stopped, though the carryover benefit was smaller after 8 years.

The part the headline usually omits

Anastrozole did not help women live longer.

Overall mortality was almost identical: hazard ratio 0.95, 95% confidence interval 0.84 to 1.06, p = 0.4. There was weak evidence of fewer deaths after recurrence in the hormone-receptor-positive subgroup, at 0.87 with a p-value of 0.09, which does not reach significance.

This is a practice-changing trial that reduced recurrences without demonstrating longer life over ten years. Both halves of that sentence are the finding.

What each drug costs

The side effect trade is real and runs in both directions.

Fractures were more common on anastrozole during active treatment: 451 versus 351, odds ratio 1.33, p below 0.0001. After treatment ended the rates were nearly identical, at 110 versus 112. Blocking estrogen weakens bone, and stopping the drug stops the effect.

Treatment-related serious adverse events were less common on anastrozole: 223 versus 369, odds ratio 0.57, p below 0.0001.

Other cancers were similar in total, at 425 versus 431, but distributed differently. Endometrial cancer occurred in 6 women on anastrozole and 24 on tamoxifen. Melanoma was 8 versus 19 and ovarian cancer 17 versus 28. Colorectal cancer ran the other way, at 66 versus 44, as did lung cancer at 51 versus 34.

Tamoxifen's endometrial risk is well established. The other differences come from a single trial and are counts, not conclusions.

When to get checked

Anyone taking an aromatase inhibitor should ask about a baseline bone density scan and how often it will be repeated. Joint stiffness and aching are the most common reason people stop these drugs, and they are worth reporting rather than enduring, because a switch to a different agent is often possible. Our page on bone loss after hormone therapy covers what protects bone during treatment.

Anyone taking tamoxifen should report abnormal vaginal bleeding promptly, especially after menopause, because that is the signal for endometrial change. New leg swelling or pain, or sudden breathlessness, needs urgent assessment for a clot.

For breast cancer itself, NCI says early disease often causes no symptoms, which is why screening matters. When changes appear, check with a doctor about a lump in or near the breast or under the arm, a thick or firm area, a change in the size or shape of the breast, and nipple changes or discharge.

The practical threshold is a new lump or firm area that persists through a menstrual cycle, or any new lump after menopause.

The wider picture

Projections published by the American Cancer Society give about 321,910 new female breast cancer diagnoses and 42,140 deaths in the United States in 2026. The median age at diagnosis is 64.

Sixty-four percent is found while confined to the breast, 27% after spread to nearby nodes, and 6% after distant spread. Five-year relative survival is essentially 100% for localized disease, 87.5% for regional, and 33.8% for distant, and 91.9% across all stages for women diagnosed from 2016 through 2022.

Those figures include every subtype. ATAC enrolled only postmenopausal women with early-stage disease, most of them hormone-receptor-positive. Our overview of hormone therapy covers how these drugs are used today.

What this trial cannot tell you

No overall survival advantage was shown at ten years. The benefit was in preventing recurrence, and a reader who converts "better" into "lives longer" has overread it.

The results apply to postmenopausal women only. Before menopause the ovaries are the main estrogen source, and blocking aromatase does not solve that.

The fracture finding is not a footnote. For a woman with osteoporosis or a prior fragility fracture, a 33% higher odds of fracture during treatment may outweigh a 4.3 percentage point difference in recurrence at ten years. That trade is individual and belongs in a conversation, not a guideline read alone.

And treatment here was five years of a single drug. Current practice includes sequences and extended durations that ATAC did not test.

Sources

• Cuzick J et al., Lancet Oncol 2010, ATAC 10-year analysis (NCBI E-utilities)
• NCI PDQ: Breast Cancer Treatment (Health Professional Version)
• NCI: Breast Cancer Signs and Symptoms
• NCI SEER Stat Facts: Female Breast Cancer

How this article was prepared

An AI-assisted editorial system helped prepare this page. No named medical reviewer has reviewed it unless one is listed.

See an error, old source, or unclear wording? Tell us.