ATLAS: Ten Years of Tamoxifen Superior to 5 Years

Audrey Andrews

March 2013, Highlights - Hormone Therapy


For the treatment of estrogen receptor (ER)-positive early breast cancer, 10 years of treatment with tamoxifen yields better outcomes than 5 years of treatment, according to an analysis from the international ATLAS study that was presented at the 2012 San Antonio Breast Cancer Symposium.

“With these new results, we see statistically fewer recurrences when patients take tamoxifen for 10 years rather than 5, and it is highly significant for breast cancer mortality and overall mortality,” said Richard Gray, MSc, Professor of Medical Statistics at Oxford University, United Kingdom. “Looking at the risk ratios, you find about a 30% reduction occurring after 10 years.”

Five years of tamoxifen is the current standard duration of treatment, based on a substantial reduction in the 15-year risk for recurrence and death compared with placebo. Further analysis of the ATLAS trial now shows that 10 years of treatment with the drug is even more effective.

Because tamoxifen has a “carryover effect” and remains protective for years after treatment is stopped, the main benefit of extended duration did not emerge until after 10 years, Mr Gray reported.

ATLAS enrolled 6846 women with ER-positive breast cancer between 1996 and 2005. Women who had taken tamoxifen for 5 years were then randomly assigned to continue treatment for another 5 years or to stop treatment immediately.

After 8 years of follow-up, there were 1328 breast cancer recurrences and 728 deaths after recurrence. The duration of tamoxifen treatment had little effect on recurrence or deaths for the first 9 years after diagnosis, but after 10 years the extended treatment led to a 25% lower recurrence rate (P = .002) and a 29% lower breast cancer mortality rate (P = .01) compared with stopping treatment at 5 years, Mr Gray said.

When 10 years of treatment with tamoxifen was compared with placebo from the start, mortality from breast cancer was essentially halved, he added.

The absolute mortality gain was 12%, or 1 life saved for every 8 women, Mr Gray noted.

Peter Ravdin, MD, PhD, Director of the Breast Health Clinic at the Cancer Therapy and Research Center of the University of Texas Health Science Center, San Antonio, commented on the findings. He noted that the results of ATLAS are most relevant to premenopausal women, because they typically receive tamoxifen and not aromatase inhibitors.

In this patient subset, “we usually stop tamoxifen at 5 years, but now we will be telling them there is evidence that 10 years is superior to 5. I am going to be comfortable doing that,” he said.

Although treatment with tamoxifen is not without its risk—0.4% of women died of endometrial cancer after 10 years of treatment with tamoxifen in the trial—the risk is “so low that it is statistically hard to discern in trials of younger women,” Dr Ravdin pointed out. “But it is always important to weigh risks and benefits.”

Therefore, for women at minimal risk of relapse—such as those with small, grade 1 tumors—it is rational to stop tamoxifen after 5 years of use, he maintained. “But some women will have difficulty with the idea of continuing beyond 5 years, and, in many cases, the amount of benefit is small enough that discontinuation is the right decision,” Dr Ravdin noted.

“Women with a relatively high risk of relapse, such as those with positive nodes and bigger tumors, will definitely be strong candidates for continuation of therapy,” he said.