Benefit of Extended Adjuvant Aromatase Inhibition in Breast Cancer Now in Question

April 2017, Vol 8, No 2

San Antonio, TX—Extended adjuvant endocrine therapy beyond 5 years with an aromatase inhibitor (AI) failed to improve disease-free survival (DFS) in patients with hormone receptor (HR)-positive breast cancer who were enrolled in the 3 large National Surgical Adjuvant Breast and Bowel Project (NSABP) B-52, IDEAL, and DATA studies, which were presented at the 2016 San Antonio Breast Cancer Symposium.

The results were surprising, because they differed from that of the large phase 3 National Cancer Institute of Canada MA.17R study, which previously showed a benefit for extending adjuvant letrozole (Femara) beyond 5 years (Goss PE, et al. J Clin Oncol. 2016;34[suppl]:Abstract LBA1).

These new results raise the question of which patients, if any, should be treated beyond 5 years of hormonal therapy in this setting. Many patients, including participants and nonparticipants in clinical trials, do not tolerate AIs well and stop taking them before their 5 years of treatment has ended. It is often difficult to convince patients to take another 5 years of AI therapy after they have completed their initial 5-year course of therapy.

Formal discussant of the 3 studies, Michael Gnant, MD, FACS, Head, Department of Surgery, Medical University of Vienna, Austria, and President, Austrian Breast and Colorectal Cancer Study Group, stated, “The trials did not reach the necessary statistical levels to demonstrate a clear benefit for aromatase inhibition extension. Extending aromatase inhibitors may be a good idea for many patients after 2 to 5 years of tamoxifen, but after initial treatment with an aromatase inhibitor, the benefits and risks must be carefully balanced on an individual basis. We are going to need the ‘art of medicine’ here.”

Compelling Concept

Despite the benefits of adjuvant endocrine therapy in early HR-positive breast cancer, approximately 50% of all patients have recurrent disease after the first 5 years of treatment (ie, late relapse). Thus, “extending adjuvant endocrine therapy beyond 5 years is a compelling concept,” Dr Gnant said.

The MA.17R study showed that women who took letrozole for a total of 10 years (beyond 5 years of tamoxifen or AI therapies) had a 34% lower risk for breast cancer recurrence and fewer contralateral breast cancers, although no overall survival (OS) benefit was found.

The findings of 3 new trials presented at the meeting were different. The results of the large NRG Oncology/NSABP B-42 trial of 3966 patients, which were presented by Eleftherios P. Mamounas, MD, MPH, FACS, Breast Surgical Oncology, UF Health Cancer Center-Orlando Health, FL, showed that 5 years of letrozole therapy after 5 years of endocrine treatment did not significantly improve DFS or OS, but did significantly improve the breast cancer–free interval and distant recurrence rates at a median of 6.9 years of follow-up.

Extended treatment with letrozole did have some benefits versus placebo on distant recurrence (3.9% vs 5.8%, respectively; hazard ratio [HR], 0.72; P = .03), new contralateral breast cancer (1.5% vs 3.0%, respectively; HR, 0.71; P = .003), and breast cancer–free interval events (6.7% vs 10.0%, respectively; HR, 0.71; P = .003).

The DATA Study

The phase 3 DATA study included 1912 postmenopausal women who received 2 to 3 years of adjuvant tamoxifen (Nolvadex) and randomized them to either 3 or 6 years of anastrozole. The results of the study, which essentially compared 5 years versus 8 years of AI therapy, were reported by Vivianne Tjan-Heijnen, MD, PhD, Maastricht University Medical Centre, the Netherlands.

The study failed to show a benefit in 5-year DFS for either 3 or 6 extra years of anastrozole treatment.

The IDEAL Study

Erik J. Blok, MD, Department of Medical Oncology, Leiden University Medical Center, the Netherlands, presented the results of the phase 3 IDEAL trial, which showed that extending letrozole to 2.5 or 5 years beyond 5 years of endocrine therapy failed to improve DFS in 1824 patients.

There was no significant difference in 5-year DFS or OS for the patients who received an additional 2.5 or 5 years of letrozole treatment. Extended treatment with letrozole significantly reduced the incidence of second primary breast cancers (HR, 0.37; P = .008), but the absolute reduction was only 1%.

Expert Commentary

Dr Gnant summarized the findings of the B-52, IDEAL, and DATA trials and provided some guidance on how to sift through these confusing data to determine the best treatment for patients.

The B-52 trial asked whether extended treatment with an AI after the initial AI treatment is of benefit. “With only a modest difference that translated to a 3.4% absolute PFS [progression-free survival] difference at 7 years, without any difference in overall survival, surely there will be debate as to whether this is clinically relevant,” Dr Gnant said.

Based on the IDEAL and DATA trial findings, there may be some benefit to extended endocrine therapy when tamoxifen is the initial treatment. In DATA, the initial therapy was tamoxifen, whereas in IDEAL, the patients received tamoxifen, an AI, or tamoxifen followed by an AI.

“Both are technically negative in their primary end point analysis,” he said, although extended treatment after initial tamoxifen produced a 21% nonsignificant relative progression-free survival benefit. Since these trials are otherwise comparable, this adds to the notion that longer duration of an AI may mainly benefit patients who started on tamoxifen.”

IDEAL and DATA exploratory subgroup analyses also suggested that patients with high-risk features may benefit the most from extending AI treatment beyond 5 years, and that tumors expressing estrogen and progesterone receptors may drive the benefits. Genomic classifiers could eventually prove helpful in selecting patients, Dr Gnant said.

Any benefit from extended AI therapy comes at a price, he emphasized. In all 3 of the trials, treatment adherence was suboptimal.

“Just 55% to 65% of patients were still on their medications after 5 years. We all know that arthralgias, myalgias, bone pain, and hot flashes are frequent with aromatase inhibitors, and compliance is likely to be even poorer outside of clinical trials….There is also likely to be a positive selection of patients in the trials who are already tolerating the drugs reasonably well,” Dr Gnant told listeners.

Based on these findings, he emphasized the importance of careful assessment of the potential risks and benefits before recommending extended treatment and individualizing decisions.

In Dr Gnant’s view, patients who are initially receiving tamoxifen could receive an AI, unless there are contraindications, such as poor bone health. The more complex scenario is for the patient to receive initial or sequential AI therapy.

“There are factors favoring extension, and others that speak against such,” he said.

Patient characteristics are key, including tolerability during the first 5 years of endocrine therapy. “Given the real risk for fractures, bone health is important and should be prominently considered. Younger age (long life expectancy) is a factor, and classical risk factors as well as tumor luminality have been proposed as factors favoring extension. Whether we use genomic assays is undecided at this point,” Dr Gnant noted.

But some patients do not tolerate AI therapy well. “Some patients actually count down the weeks and days until they can stop. It is unlikely we can convince them to go on extended treatment, except in cases of exceptional risk,” he said. Patients who have a low risk for recurrence according to clinical and genomic assessments are likely also not candidates for extended treatment.

Dr Gnant said that “the mystery of chronic residual risk” in HR-positive breast cancer is not well understood. “We need new concepts, such as directly targeting stem cells to eliminate the source of late metastases for good, to attack the chronic part of luminal breast cancer recurrence risk. Attacking the enemy with our ordinary drugs apparently will not cure all patients,” he concluded.

Related Articles