e18581 Background: While most patients (pts) with HR+, HER2- early breast cancer (EBC) do not experience disease recurrence, those with high risk clinicopathologic features may have recurrence within the first few years on adjuvant endocrine therapy (AET). This study estimates risk of recurrence in pts with EBC based on clinicopathologic features and evaluates the medical need for more efficacious treatments using US oncology practice data. Methods: This retrospective study used the nationwide Flatiron Health electronic health record derived deidentified database. The cohort included pts with HR+, HER2- stage I-III breast cancer, diagnosed Jan 2011-Mar 2020, who received surgery and AET. Clinicopathologic features were used to identify a ‘high risk (HiR) group’ (≥ 4 positive axillary lymph nodes (LN), or 1-3 positive axillary LN and ≥ 1 of the following: Grade 3, tumor size ≥ 5 cm, or Ki-67 ≥ 20%) and ‘low risk (LoR) group’ (pts who do not meet above criteria, including a subset with node negative (N0) disease). Cox proportional hazards regression models compared invasive disease-free survival (IDFS) and distant relapse-free survival (DRFS) from AET initiation between groups according to a sequential gatekeeping strategy and stepwise analysis: first compare HiR group to N0 pts; if significant then compare HiR to LoR group. Results: 557 (13.8%) pts were in the HiR group and 3,471 (86.2%) in the LoR group (including 2,867 N0 pts). The study population (median age 64 yrs) was predominantly female (99.2%), white (69.4%) and postmenopausal (75.6%) with median follow-up of 39.6 months. Pts in the HiR group were younger and more likely premenopausal, have a BRCA mutation, invasive lobular carcinoma and less likely to have received an Oncotype DX test compared to LoR pts. Of the 557 HiR pts, 231 (41.5%) had ≥ 4 positive LN and 326 (58.5%) had 1-3 positive axillary LN with additional risk factor(s): tumor size ≥ 5 cm (11.7%), histologic grade 3 (32.0%), and/or Ki-67 ≥ 20% (31.6%). Most HiR pts received radiotherapy (82.4%) and chemotherapy (68.1%). Significant differences in IDFS and DRFS were observed between HiR group and N0 and LoR groups (logrank test p < .0001 for all). The 2-year IDFS rate was 88.1% in the HiR group, compared to 97.4% in N0 pts and 97.1% in the LoR group; 2-year DRFS rates were 89.0% compared to 97.9% and 97.7%, respectively. Pts in the HiR group had significantly higher hazard of invasive disease recurrence or death compared to N0 (HR = 3.42, 95% CI: 2.69-4.34, p < .0001) and LoR group (HR = 3.07, 95% CI: 2.46-3.84, p < .0001). Similar results observed for DRFS (HiR vs N0: HR = 3.46, 95% CI: 2.70-4.44, p < .0001; HiR vs LoR HR = 3.16, 95% CI: 2.50-3.98, p < .0001). Conclusions: Approximately 12% of pts with EBC and high risk clinicopathologic features experienced invasive disease recurrence or death within 2 years of initiating AET. Optimal use of standard therapies and novel treatment options are needed to prevent early recurrence and metastases in these pts.
Serum metabolomic profiles evaluated after surgery may identify patients with oestrogen receptor negative early breast cancer at increased risk of disease recurrence. Results from a retrospective study