Recurrence risk stratification by Dutch clinical risk criteria for early-stage luminal breast cancer patients in Taiwan: A population-based analysis.

Abstract

e12509 Background: Dutch clinical risk criteria (low-risk definition: age > 35 years and (grade 1 with tumor ≤3cm, grade 2 with tumor ≤2cm, or grade 3 with tumor ≤1cm) have been used to stratify the benefit of MammaPrint and Oncotype DX for the decision-making regarding adjuvant chemotherapy for early-stage luminal breast cancer. We propose that the criteria could help to identify low-risk patients who could barely benefit from multi-gene testing. Methods: Breast cancer patients from Taiwan Cancer Database initially treated with primary surgeries between 2008 and 2012 who met the following criteria: (1) pathologic node-negative, (2) hormone receptor-positive, (3) HER2-negative, (4) undergone hormonal therapy, and (5) a minimum follow-up time of 5-year if free from any event, were enrolled in this study. Out of the total 2679 eligible patients, 1074 (40.1%) patients received adjuvant chemotherapy in addition to endocrine therapy. The study endpoints included breast cancer-specific survival (BCSS) and overall survival (OS). Kaplan-Meier statistics estimated the difference between clinical outcomes in low- and high-risk groups. Results: The median follow-up time of BSCC and OS was 5.9 years (range, 0-7 years) and 5.8 years (range, 0-7 years), respectively. There were statistical significances of 5-year BCSS (n = 2679) and 5-year OS (n = 2636) between low-risk and high-risk groups (in both endpoints, P < 0.0001). According to the Dutch criteria, low-risk patients with and without adjuvant chemotherapy had a 5-year BCSS of 99.0% vs. 99.2% and a 5-year OS of 98.4% vs. 97.4%, respectively. High-risk patients with and without adjuvant chemotherapy had a 5-year BCSS of 97.7% vs. 98.1% and a 5-year OS of 96.4% vs. 95.3%, respectively. Conclusions: The benefit of chemotherapy in low-risk patients classified by Dutch criteria might be very small since the breast cancer mortality was less than 1% with a minimum of 5-year follow-up. Dutch criteria cannot identify high-risk patients who would benefit from chemotherapy. We assumed that multi-gene testing in low-risk patients would not be cost-effective.