Stratification of oncotype (ODX) Recurrence Score (RS) by Breast Cancer Index (BCI) and its relation to discordance.

Abstract

e12025 Background: ODX is a gene-expression panel predictive of 10-year risk of distant recurrence and has shown prognostic value in both node-negative (LN-) and node-positive (LN+) breast cancer (BC) patients. BCI predicts 10-year risk of recurrence and also identifies those that may benefit from extended adjuvant endocrine therapy (EET). We stratified ODX RS by BCI to compare its precision in predicting recurrence category and its relation to discordance. Methods: 49 women with HR+, LN- and LN+ BC with available ODX and BCI RS with median follow up of 6 years were included. BCI risk groups were evaluated using predetermined cutoffs. RS were stratified by traditional ODX and TAILORx cutoffs and were considered discordant if there was a 1-step or 2-step difference between ODX risk group (low, intermediate, high) and tumor grade (well, moderately and poorly differentiated). N (%) or median (range) were used to describe patients’ characteristics between BCI groups and compared by chi-square test or Fisher’s exact at the confidence level of 5%. Results: Using traditional RS cutoffs, BCI reclassified 42.3% of patients with low RS as intermediate (IR) or high risk (HR). 69% with intermediate RS were reclassified to low risk (LR) and 12% were upgraded to HR by BCI. 27% of low and intermediate RS had a high benefit from EET. 2-step discordance was 14.3%. Using TAILORx RS cutoffs, BCI reclassified 20% of patients with low RS as IR. None were upgraded to HR. In the intermediate RS group, 58% were reclassified as LR and 14% as HR. 24% of intermediate RS had a high benefit from EET. 2-step discordance was 2%. Conclusions: Patients classified as LR by TAILORx RS cutoffs are less likely to be reclassified as high or intermediate risk by BCI. No significant association was found between discordance among ODX RS and tumor grade and BCI prediction of recurrence or benefit of EET.