Recurrence score (RS) based on 21-gene expression assay is widely used to evaluate prognosis and potential benefits of adjuvant chemotherapy (AC) for hormone-receptor positive (HR+) breast cancer (BC) patients (pts). In this study, we investigated clinical significance of RS in premenopausal HR+ BC pts.
HR+ BC pts with age < 50 years at diagnosis who had at least 3 years of follow-up duration from surgery and had 21-gene expression assay in Asan Medical Center, Seoul, Korea, between June 2005 to July 2018 were included. Recurrence-free survival (RFS) by STEEP version 2.0 was compared according to the RS and other clinicopathologic variables.
Among total 567 pts included in this analysis, 117 pts (20.6%) had age < 40 years and 238 pts (42.0%) had luminal B type BC (LB) by 4-IHC, 85 pts (15.0%) had RS > 25, 301 pts (53.8%) had high clinical risk BC as defined by tumor size, lymph node and histologic grade and 92 pts (16.2%) had AC. On univariate analysis, pts with RS > 25 showed poor RFS (log-rank, p = 0.007). Also, pts with age < 40 years (log-rank, p = 0.008), high clinical risk (log-rank, p = 0.030) and LB (log-rank, p = 0.002) had poor RFS. On multivariate analysis, age < 40 years and LB were associated with significantly poor RFS with HR of 2.49 (95% CI 1.16-5.36, p = 0.020) and 2.59 (95% CI 1.11-6.04, p = 0.028), respectively while RS > 25 was not associated with RFS. When compared according to age group (< 40 years vs. 40 years) and luminal types (luminal A (LA) vs. LB), pts with age < 40 years and LB showed significantly poor RFS (HR 4.52, [95% CI 2.08-9.84], p < 0.001) compared to pts with age 40 years or LA. Pts with < 40 years with LB had higher proportion of pts who received AC (35.0% vs. 14.0%, p < 0.001) compared to pts with age 40 years or LA. While RS > 25 was associated with poor RFS among pts with age 40 years or LA (HR 3.20, [95% CI 1.20-8.53], p = 0.020), RS > 25 was not associated with RFS in pts with age < 40 years and LB (HR 1.31, [95% CI 0.34-5.10], p = 0.699).
The RS alone was not prognostic among premenopausal HR+ BC pts especially with age < 40 years and LB, which showed significantly poor RFS compared to pts with age 40 years or LA despite of higher proportion of pts who received AC.
Clinical trial identification
Legal entity responsible for the study
Has not received any funding.
All authors have declared no conflicts of interest.