Oncotype Recurrence Score (RS) at the Extremes of Tumor Size: Which Drives Clinical Outcomes?

Abstract

Tumor size is an established and independent risk factor for local-regional recurrence (LRR) and distant recurrence (DM). More recently, the recurrence score calculated from a 21-gene expression assay (Oncotype DXTM, Exact Sciences) has also been correlated with LRR and DM. We sought to determine the impact of the interaction between tumor size and Oncotype RS, particularly when the variables are in discrepancy with each other. More specifically, we evaluated clinical outcomes in patients with small tumors (≤1 cm) and high RS (≥26) and, separately, in patients with large tumors (>5 cm) with low RS. Between 2008 and 2020, 310 patients were identified retrospectively as having been treated for early-stage, hormone-receptor positive, Her2-negative breast cancers with tumor size ≤1 cm that were node-negative and had RS ≥ 26 at our institution. In addition, 64 patients were identified with tumor size >5 cm and RS < 26 (irrespective of nodal status). Locoregional recurrence rates (LRR) and invasive recurrence rates (composite of LRR and distant recurrence) were estimated using the Kaplan-Meier method. Patient characteristics are shown in Table 1. In the group of patients with small tumors and high RS, the 5- and 10-year invasive recurrence rates with 95% CI were 8% (4.2-12) and 17% (8.2-26). The 5- and 10-year locoregional recurrence rates with 95% CI were 5.8% (2.7-8.8) and 15% (6.2-23). In the group of patients with large tumors and low RS, 10-year rates could not be estimated with the available data. The 5- and 8-year invasive recurrence rates with 95% CI were 3.2% (0-7.5) and 3.2% (0-7.5). The 5- and 8-year locoregional recurrence rates with 95% CI were 1.6% (0-4.7) and 1.6% (0-4.7). Our findings suggest that patients with small tumors and high RS are at a higher risk for LRR compared to the average ≤1 cm node-negative breast cancer based on published data on the effect of tumor size on LRR and DM. Similarly, tumors larger than 5 cm with low RS appear to behave indolently and in a manner consistent with more favorable risk (despite their large size). These findings may have important implications for the tailoring of local-regional treatment strategies.