ATLAS

Published

Davies C, et al. "Long-term effects of continuing adjuvant tamoxifen to 10 years versus stopping at 5 years after diagnosis of oestrogen receptor-positive breast cancer: ATLAS, a randomised trial". The Lancet. 2013. 381(9869):805-16.
PubMed • Full text • PDF

Clinical Question

Among women with breast cancer, how does 10 years of adjuvant tamoxifen compare to 5 years of tamoxifen in reducing breast cancer recurrence and mortality?

Bottom Line

Among women with breast cancer, 10 years of adjuvant tamoxifen reduces all-cause mortality and breast cancer recurrence compared to 5 years of therapy, at the expense of marginally increased rates of endometrial cancer and VTE.

Major Points

Two or five years of adjuvant tamoxifen had been the standard of care among women with estrogen receptor (ER)-positive breast cancer. Whether extension of therapy to 10 years or beyond results in improved efficacy and an acceptable side effect profile was and remains incompletely understood.

The 2013 Adjuvant Tamoxifen: Longer Against Shorter (ATLAS) trial randomized 15,244 women with breast cancer who had completed about 5 years of adjuvant tamoxifen to either 1) continued tamoxifen to 10 years, or 2) discontinuation of tamoxifen at 5 years. Approximately half of patients (53%) were ER-positive and were included in the efficacy analysis; those with ER-negative or ER-unknown disease were included in analyses of non-breast-cancer outcomes. With a mean follow-up of 7.6 years, extended tamoxifen was associated with a small but statistically significant reduction in all-cause mortality (21.5% vs. 22.9%), at the expense of more endometrial cancer (1.8% vs. 0.97%), pulmonary embolism (0.63% vs. 0.33%). Incident CAD was reduced (2.01% vs. 2.53%). Among those with ER-positive disease, there was a more pronounced difference in all-cause mortality (18.6% vs. 21.1%) and breast cancer recurrence (18% vs. 21%).

ATLAS provides evidence that continued therapy to 10 years provides modest benefit for some women with ER-positive breast cancer. These findings were redemonstrated in an interim analysis of the aTTOM trial, presented at the 2013 ASCO Annual Meeting. Whether the benefit of extended tamoxifen applies to aromatase inhibitors is not clear; the NSABP B-42 and MA.17R trials are underway and will help to address the role of extended letrozole for postmenopausal patients. Whether extended endocrine therapy should be offered to all patients remains unknown, and the NCCN guidelines merely recommend to consider continuation of endocrine therapy following 5 years of initial treatment.

Guidelines

NCCN Breast Cancer (3.2013)

• For women premenopausal at the time of diagnosis, tamoxifen for 5 years (category 1) ± ovarian suppression or ablation (category 2B)
• If postmenopausal after 5 years of initial tamoxifen, then consider tamoxifen for additional 5 years (category 1) or switch to aromatase inhibitor for additional 5 years (category 1)
• If premenopausal after 5 years of initial tamoxifen, then consider tamoxifen for additional 5 years (category 1) or pursue no further endocrine therapy

Design

• Randomized, open-control, multinational trial
• N=12,894
  • Continued tamoxifen to 10 years (n=6454)
  • Stop tamoxifen at 5 years (n=6440)
• Setting: 36 countries
• Enrollment: 1996-2005
• Mean follow-up: 7.6 years
• Analysis: Intention-to-treat
• Primary outcome: All-cause mortality

Population

Inclusion Criteria

• Female
• Early breast cancer
• Receiving adjuvant tamoxifen
• Clinically free of disease

Exclusion Criteria

• Contraindication to tamoxifen (not protocol-defined, but clinicians encouraged to consider pregnancy, breastfeeding, retinopathy, need for coagulation therapy, endometrial hyperplasia, low risk of death from breast cancer, and presence of other life-threatening disease)

Baseline Characteristics

From the continued tamoxifen group.

• ER status
  • ER positive: 53%
  • ER negative: 10%
  • ER unknown: 37%
• Age
  • <45 years: 19%
  • 45-54 years: 32%
  • 55-69 years: 40%
  • ≥70 years: 9%
• Node status
  • Node negative: 52%
  • N1-N3: 26%
  • ≥N4: 15%
  • NX: 7%
• Tumor diameter
  • 1-20mm: 38%
  • 21-50mm: 43%
  • >50mm: 10%
  • Unknown: 10%
• Prior duration of tamoxifen
  • 4-4.9 years: 33%
  • 5-5.9 years: 57%
  • ≥6 years: 10%
• Hysterectomy: 17%
• Postmenopausal: 90%
• Site of enrollment
  • Europe, Australia, NZ, USA, South Africa: 39%
  • Latin America: 27%
  • Asia/Middle East: 34%

Interventions

• Randomized to continued tamoxifen (up to 10 years total) or discontinuation of tamoxifen (after 5 years total).
• Tamoxifen regimen was Nolvadex 20mg daily.
• Events coded by principal investigator, not masked to treatment allocation.

Outcomes

Comparisons are 10 years vs. 5 years of total tamoxifen exposure.

Primary Outcome

Death, any ER status

21.5% vs. 22.9% (RR 0.93; 95% CI 0.86-1.00; P=0.04)

Secondary Outcomes

Death, ER-positive women

18.6% vs. 21.1% (RR 0.87; 95% CI 0.78-0.97; P=0.01)

Breast cancer recurrence, ER-positive women

18.0% vs. 20.8% (RR 0.84; 95% CI 0.76-0.94; P=0.002)

Adverse Events

Endometrial cancer

1.80% vs. 0.97% (RR 1.74; 95% CI 1.30-2.34; P=0.0002)

Pulmonary embolism

0.63% vs. 0.33% (RR 1.87; 95% CI 1.13-3.07; P=0.01)

Stroke

2.01% vs. 1.48% (RR 1.06; 95% CI 0.83-1.36; P=0.63)

CAD

2.01% vs. 2.53% (RR 0.76; 95% CI 0.60-0.95; P=0.02)

Criticisms

• The authors did not stratify patients according to initial treatment strategy (eg, surgery, radiation, and chemotherapy) and therefore potential interactions between initial treatment strategy and duration of endocrine therapy were not available in the ATLAS data set.
• The magnitude of benefit was quite small (absolute risk reduction for all-cause mortality: 0.014) and of borderline statistical significance (P=0.04), which raises questions about the clinical benefit likely to be observed by extending tamoxifen use to 10 years.

Funding

Sponsored by Oxford University.