This article is more than 5 years old. Information may no longer be current.
Meta-analysis reveals significant interaction between HER-2 status, outcomes with aromatase inhibitors
Click Here to Manage Email Alerts
SAN ANTONIO — A significant interaction exists between HER-2 status and treatment of aromatase inhibitors compared with tamoxifen, according to the results of a meta-analysis presented at San Antonio Breast Cancer Symposium.
Aromatase inhibitors improved outcomes among patients with HER-2–negative breast cancer, whereas they offered no benefit — and may have had a negative effect — on patients with HER-2–positive disease.
HER-2 has long been proposed as a marker of endocrine resistance, and results from three prior trials — all of which were conducted prior to utilization of HER-targeted therapy — suggested a potential role for HER-2 in the selection of patients for upfront treatment with aromatase inhibitors. Despite these findings, each trial lacked the power to assess treatment by HER-2 interaction due to small sample sizes and limited numbers of HER-2–positive patients.
“We proposed that a meta-analysis of effects of HER-2 — specifically within the first 3 years of endocrine therapy — may have the potential to inform patient selection for either upfront aromatase inhibitor treatment or a switch strategy of tamoxifen followed by aromatase inhibitors, allowing for more extended aromatase inhibitors with lesser side effects,” John M.S. Bartlett, PhD, director of transformative pathology at Ontario Institute for Cancer Research in Toronto, said during a presentation.
Bartlett and colleagues with the Translational Aromatase Inhibitors Overview Group performed a meta-analysis of the randomized phase 3 ATAC, BIG 1-98 and TEAM trials, which comprised a combined 12,129 patients with centrally confirmed ER and HER-2 status.
Researchers obtained patient-level data from the Cancer Research UK Clinical Trials Unit to compare aromatase inhibitors with tamoxifen during the first 2 to 3 years of adjuvant treatment. The analysis accounted for the established time-dependency of relapse in HER-2–positive patients who did not receive endocrine therapy. It also was designed to compare upfront aromatase inhibitor treatment vs. sequential strategies.
Investigators used covariate-adjusted Cox models for each trial to estimate how the interaction between HER-2 status and treatment affected distant recurrence-free interval.
Bartlett and colleagues performed a meta-analysis of interaction between HER-2 status and treatment effects.
Distant recurrence-free interval up until the pre-planned 2- to 3-year treatment switch time for those trials with a switching element served as the primary endpoint. Patients who remained alive and in follow-up with no evidence of disease were censored at 2 to 2.75 years.
The meta-analysis included 12,129 patients, or 97.4% of the cases required to achieve 90% power to detect a treatment-biomarker interaction.
Of these patients, 1,092 (9%) were HER-2 positive. Researchers analyzed 473 (3.9%) distant recurrences, of which 111 (23.4%) occurred in HER-2–positive patients.
After adjustments for clinical covariates such as PR status, grade and nodal status, the investigators estimated a pooled HER-2-by-treatment interaction on distant recurrence-free interval of 1.61 (95% CI, 1.01-2.57). They calculated treatment effect HRs for aromatase inhibitors vs. tamoxifen of 1.13 (95% CI, 0.75-1.71) for HER-2–positive patients and 0.7 (95% CI, 0.56-0.87) for HER-2–negative patients.
Results revealed heterogeneity among interaction terms (I2 = 59%), which was driven by treatment effect heterogeneity in the HER-2–positive group (I2 = 71%; P = .03) but not the HER-2–negative subgroup (I2 = 0%).
DFS results appeared similar between groups.
Bartlett provided several caveats to the study results, most importantly that most patients with HER-2–positive cancers were not eligible for chemotherapy and, therefore, not eligible for HER-2–directed therapies such as trastuzumab (Herceptin, Genentech).
“The impact of HER-2–directed therapies in the context of this trial is unknown,” Bartlett said during his presentation. “This alone precludes a recommendation for us to change practice for patients who are eligible for trastuzumab now as a monotherapy.”
Also, the small number of HER-2–positive cancers and events may explain much of the heterogeneity in the treatment effects within the HER-2–positive subgroups in the three trials analyzed, Bartlett said. Of the 111 events that occurred in the HER-2–positive population, more than 50% occurred in one trial.
“This suggests that some caution should be applied before we interpret this into clinical practice,” Bartlett said. – by Mark Leiser
Reference:
Bartlett JMS, et al. Abstract S4-06. Presented at: San Antonio Breast Cancer Symposium; Dec. 8-12, 2015; San Antonio.
Disclosure: Bartlett reports honoraria for consultant work from Insight Genetics and BioNTech, as well as three pending patents. Please see the abstract for a full list of all researchers’ relevant financial disclosures.