br Acknowledgements This work was supported by NIH HL JHF

Acknowledgements This work was supported by NIH HL111674 (JHF), and the Antoinette E. (“Mimi”) & Herman Boehm Foundation (JK). The authors wish to acknowledge the valuable assistance of the Functional Genomics Core of the University of Colorado, Denver.
Introduction The 21-gene recurrence score (RS) assay (Oncotype DX; Genomic Health Inc, Redwood City, CA) is a quantitative reverse-transcriptase polymerase chain reaction assay to assess the expression of 21 cancer-related genes in patients with early stage breast cancer. The RS has been extensively validated as a tool to measure risk of distant recurrence and the benefit of adjuvant chemotherapy in lymph node (LN)-negative and LN-positive patients with SMIP004 receptor-positive (ER+), human epidermal growth factor receptor-2-negative (HER2−) breast cancer.1, 2, 3, 4, 5, 6, 7 Currently, RS testing is recommended by the National Comprehensive Cancer Network (NCCN) guidelines for patients with LN-negative, ER+, HER2− breast cancer. HER2 is a growth factor receptor gene that is amplified in 25% to 30% of breast cancer cases. Overexpression of HER2 has been shown to impact tumor biology, causing a more aggressive disease as manifest by worse disease-free and overall survival.9, 10, 11, 12 Owing to the aggressive nature of HER2-positive (HER2+) breast cancer, guidelines recommend the addition of adjuvant chemotherapy and targeted HER2 therapy with trastuzumab. The initial validation studies of RS testing were performed in patients with HER2+ and HER2− disease; however, all subsequent studies have excluded patients with HER2+ disease, likely owing to their worse prognosis. We sought to evaluate rates of RS testing, to define the distribution of RS, and to determine the impact of RS results on clinical decision-making for patients with ER+, HER2+ breast cancer.
Patients and Methods An augmented version of the Surveillance, Epidemiology, and End Results (SEER) program database was used to identify women with ER+, HER2+, stage I and II carcinoma of the breast from 2010 to 2013, with linked data on RS testing. The SEER program database is an aggregate of 18 distinct geographic cancer registries that represents 28% of the United States population. Women with ER-negative (ER−) or unknown status were excluded. The study was exempt from institutional review board review at the University of Minnesota as only deidentified data was used.
Results We identified 17,101 patients with HER2+, ER+, stage I/II breast cancer in the SEER database from 2010 to 2013. Of those, 59% of patients were age 55 or older, 78% of patients were white, and 85% of patients had private insurance. Most tumors were smaller than 2 cm (53%), LN-negative (65%), progesterone receptor-positive (PR+) (72%), and grade I/II (49%). The cohort was equally distributed between stage I and stage II disease (Table 1). Of the 17,101 patients, 1081 (5%) patients underwent RS testing. Factors associated with a decreased odds of RS testing were younger patient age, other race (compared with non-Hispanic white), more recent year of diagnosis (2012-2013 compared with 2010-2011), Medicaid insurance coverage (compared with private), positive or unknown LN status (compared with negative), PR-negative (PR−) or unknown status (compared with PR+), larger tumor size, and grade III tumors (compared with grade I/II) (Table 2). The distribution of RS results by TAILORx cutoffs were: high risk, 17%; intermediate risk, 49%; and low risk, 34%. Factors significantly associated with high-risk RS were PR− status, tumor size greater than 2 cm, and grade III tumors (Table 3).
Discussion The 21-gene RS test is validated in patients with ER+, HER2− early stage breast cancer. Using a large population-based dataset, we found that RS testing was being performed in 5% of patients with HER2+ disease. Furthermore, in this subset of HER2+, ER+ patients, we found that increased age led to an increased odds of RS testing, whereas PR− status and high-grade tumors were less likely to be tested. We also found that PR− status, tumor size greater than 2 cm, and grade III tumors were significantly associated with an increased odds of a high-risk TAILORx score. Chemotherapy utilization was greater in patients with a high-risk score than in patients with a low or intermediate score. However, rates of chemotherapy utilization among patients with a high-risk RS were comparable with rates in patients that were not tested.