Carlson, Robert H.
SAN ANTONIO—The eagerly awaited results of the ATLAS trial show that 10 years of hormonal therapy with tamoxifen have a carryover benefit to at least 15 years after initiation of treatment.
Previous studies have shown that five years of treatment are associated with a one-third reduction in recurrence and in breast cancer mortality in years five to 10 compared with no tamoxifen.
And now the ATLAS (Adjuvant Tamoxifen—Longer Against Shorter) researchers reported at the CTRC-AACR San Antonio Breast Cancer Symposium that this carryover effect is even stronger after 10 years of treatment. The results were published simultaneously online ahead of print in The Lancet (doi:10.1016/S0140-673661963-1).
At 15 years, the recurrence rate for women treated with tamoxifen for only five years was 25.1 percent vs. 21.4 percent for women who received tamoxifen for 10 years. And the rate of breast cancer mortality for women treated for five years was 15.0 percent vs. about 12 percent for women who received tamoxifen for 10 years.
RICHARD GRAY, MSC: T...Image Tools
That represents an absolute gain of 2.8 percent favoring continuing treatment with tamoxifen for 10 years, said Richard Gray, MSc, Professor of Medical Statistics at the University of Oxford, who presented the data here on behalf of the ATLAS investigators. First study author was Christina Davies, MD, a coordinator in the Clinical Trial Service Unit at the University of Oxford.
A previously published meta-analysis in The Lancet (2011;378:771-784) showed that five years of tamoxifen reduces breast cancer recurrence in the first decade (years 0-9), with little further gain later, and reduces breast cancer mortality substantially throughout years 0-14.
In ATLAS, 6,846 women in 36 countries with ER-positive disease who had five years of adjuvant tamoxifen were randomly assigned between 1996 and 2005 to continue another five years (to year 10) or stop at year five. Half had node-positive disease. Annual follow-ups recorded compliance, hospital admissions, breast cancer recurrence (including new contralateral), any other new primary cancer, and cause of death.
At 15 years there were 617 recurrences in the 10-year treatment group vs. 711 in the five-year group. There were 331 deaths due to breast cancer in the 10-year treatment group vs. 397 in the five-year group. And deaths from all causes were 639 vs. 722.
Gray noted that continuing tamoxifen use can increase side effects, with endometrial cancer being the most life-threatening. But he said the absolute mortality gain of 10 years of tamoxifen therapy at 15 years was 12 percent—one woman in eight—vs. a 0.4 percent endometrial cancer mortality loss, about one woman in 250.
“The reduction in breast cancer deaths far outweighs the risk of endometrial cancer and other side effects of tamoxifen,” Gray said. And in premenopausal women, for whom tamoxifen is often the endocrine treatment of choice, there was no apparent excess of endometrial cancer.
The study was funded by Cancer Research UK, the UK Medical Research Council, AstraZeneca, the United States Army, and the European Union.
Large Public Health Implications
Study Discussant WIL...Image Tools
The small difference of two to three percent in mortality shown in ATLAS could have very large public health implications, “because this is an inexpensive drug and could be applied worldwide,” said the Discussant for the study at the meeting, William E. Barlow, PhD, Research Professor in the Department of Biostatistics at the University of Washington and Lead Statistician for the SWOG Breast Committee and Member of the Clinical Research Division of Fred Hutchinson Cancer Research Center.
Barlow noted that while the ATLAS paper focuses on recurrence and breast cancer mortality in ER-positive disease, the primary outcome was actually all-cause mortality—“which was also statistically significant, and was almost buried in the manuscript.”
But the big, unanswered question, he said, is how will these results compare with switching women after five years of tamoxifen to an aromatase inhibitor.
Barlow said that unfortunately, ATLAS can't be directly compared with the randomized, Phase III MA.17 trial, which found that five years of letrozole after discontinuation of tamoxifen extended disease-free survival and, for some patients, overall survival, due to different outcome measures and duration of follow-up. “But in general, the disease-free survival and recurrence rates [of MA.17] would tend to favor switching to the aromatase inhibitor letrozole rather than continuing with tamoxifen,” he said.
Nevertheless, he said, patients who are still premenopausal without ovarian suppression or those who are not strong candidates for aromatase inhibitors would likely benefit from extended tamoxifen despite the increased risk from endometrial cancer.
PETER RAVDIN, MD, PH...Image Tools
The result of the ATLAS trial will have an immediate impact on the treatment of premenopausal women, said SABCS Co-Director Peter Ravdin, MD, PhD, SABCS Co-Chair and Director of the Breast Health Clinic at the Cancer Therapy & Research Center at The University of Texas Health Science Center at San Antonio.
Tamoxifen is the primary hormonal therapy for premenopausal women because aromatase inhibitors do not work for those patients, he noted. In postmenopausal women with early breast cancer, aromatase inhibitors are the standard of care.
“For women approaching five years of therapy, we have usually been telling them at this point that we would stop tamoxifen, but now we will be telling them there is clinical evidence that 10 years is superior to five years,” he said at a news conference at the meeting. “And I'm going to be comfortable doing that. Endometrial cancer is a risk, but the risk is very low in premenopausal women, in fact so low that in these clinical trials it is statistically hard to tell whether or not tamoxifen increases the risk of endometrial cancer in women who are still premenopausal.”
Ravdin said that in the biology of breast cancer, some women are fated to have late relapses that current early therapies aren't effective at blocking.
“But we can do better by treating women with hormonal therapy beyond five years,” he said. Women who have relatively high risks of late relapse—with positive nodes or larger tumors, for example—would be strong candidates for continuation of therapy, Ravdin said.
“But a patient with a small Grade 1 tumor and a very low risk of both early and late relapse may rationally decide she doesn't want to take tamoxifen beyond five years. An additional one-third reduction in a very small risk is still a very small benefit.”
© 2013 Lippincott Williams & Wilkins, Inc.