SAN ANTONIO – Five years of additional therapy with anastrozole (Arimidex) was no more effective than 2 additional years following the standard 5 years of initial endocrine therapy in postmenopausal women with hormone receptor–positive (HR+) breast cancer, Austrian investigators reported.
In fact, the only thing that the additional years of the aromatase inhibitor (AI) anastrozole seemed to add was increased risk for fracture, said Michael Gnant, MD, from the Medical University of Vienna, on behalf of colleagues in the ABCSG-16 trial.
Dr. Michael Gnant
Previous trials have convincingly demonstrated the benefit of giving patients an AI for 5 years after 5 years of tamoxifen, but the optimal duration of extended adjuvant AI therapy is not known, Dr. Gnant said.
The ABCSG trialists recruited 3,484 postmenopausal women from with HR+, stage T1-3, node-negative or -positive, nonmetastatic breast cancer who had completed 4-6 years of endocrine therapy with either tamoxifen, an AI, or tamoxifen followed by an AI. The patients were randomly assigned at the end of initial endocrine therapy to either 2 years or 5 years of anastrozole.
As noted before, disease-free survival (DFS), the primary endpoint, was virtually identical between the treatment arms. The DFS rate at a median of 8.75 years after randomization – that is, approximately 14 years after diagnosis – was 71.1% among patients treated for 2 additional years, vs. 70.3% for patients treated for 5 extra years, translating into a hazard ratio of 1.007 and making the contest a statistical dead heat.
Similarly, there was no difference by anastrozole duration in the secondary endpoint of overall survival at 10 years, with respective rates of 85.3% vs. 84.9%, with a hazard ratio identical to that in the DFS analysis.
Where the 5-year schedule surpassed the 2-year schedule, however, was in apparent risk for fractures, which was 6.3% after 5 years of additional therapy, compared with 4.7% at 5 years among patients who received just 2 additional years of anastrozole. The hazard ratio associated with the difference was 1.353, but because the lower end of the 95% confidence interval was 1.00, the finding was of borderline significance (P = .053), Dr. Gnant acknowledged.
There are several ongoing translational studies that may help to identify specific molecular characteristics that could predict benefit from prolonged extended therapy in a given patient, “but for now we can conclude that 7 years are good enough for almost every patient with luminal breast cancer,” Dr. Gnant said at a briefing prior to his presentation of the study in an oral session.
“I do believe that for us as clinical scientists a negative trial is always disappointing, but I think the clinical take-home message can actually help to avoid unnecessary side effects for many, many women,” he added.
Asked at the briefing whether, given the identical survival curves between the two trial arms, additional therapy beyond 5 years was needed, Dr. Gnant replied “that was addressed by other trials. I think that the trials after tamoxifen are very clear: We have hazard ratios around 0.6 after tamoxifen, so some type of extension for adding aromatase inhibitors should be the standard of care.”
He noted that the optimal duration of additional therapy with an AI has not been known, because the trial that could have answered that question, the MA-17 trial, was halted and unblinded after just 2.5 years when an interim analysis showed superior survival with letrozole (Femara), compared with placebo.
More than 60% of patients in the placebo group in that trial were crossed over to letrozole, further muddying long-term follow-up results.
Carlos Arteaga, MD, director of the Harold C. Simmons Comprehensive Cancer Center at University of Texas Southwestern Medical Center in Dallas, who moderated the briefing, agreed with Dr. Gnant that this ostensibly negative trial had good results for patients.
“I hope that we continue to see more de-escalation studies. I hope that as we combine AIs with CDK4/6 inhibitors, we may make therapy even shorter. I think we should do better than just extending and extending and extending. We have to come up with better ideas,” he said in an interview.
The ABCSG-16 study was supported by AstraZeneca. Dr. Gnant disclosed research funding, honoraria, and travel funding from that company and others. Dr. Arteaga disclosed consulting fees from AstraZeneca and other companies.
SOURCE: Gnant et al. SABCS 2017 Abstract GS3-01