Assessment of method of breast cancer detection and Oncotype DX recurrence score.

Abstract

581 Background: Data indicate that screen-detected (SD) cancers show more favourable pathological characteristics and have a better prognosis compared to symptomatic or interval cancers (SY). These findings can be partly explained by length time and lead time biases. OncotypeDX recurrence score (RS) provides prognostic information independent of traditional clinicopathological variables and provides a robust measure of tumour proliferation. The objective of this study was to determine whether the RS would vary depending on the method of detection (MOD) in patients with lymph node-negative, ER-positive, HER-2 negative breast cancer (BC). Methods: We included patients with ER-positive, lymph node-negative and HER2-negative BC who underwent OncotypeDX testing as part of routine care or a clinical trial. We retrospectively reviewed patient charts and extracted information on MOD (SD versus SY), clinicopathological variables and RS. Chi-squared test was used to test for differences between categorical variables and MOD. Analysis of variance (ANOVA) was used to investigate the relationship between RS and MOD and clinicopathological variables. Results: We identified 596 patients from 4 academic hospitals (Mater University Hospital n= 98 (16%), St Vincent’s Hospital n= 161 (27%), Waterford Regional Hospital n= 27 (4%) and University of Texas, MD Anderson Cancer Center n= 310 (52%)). There were 359 (60 %) SD- detected and 237 (40%) SY-detected breast cancers. Clinicopathological variables were similar for age at diagnosis (56 versus 52 years), tumour size (16mm versus 18mm), grade (63% versus 61% grade 2; 15% versus 15% grade I; 21% versus 24% grade 3) and LVI (18% versus 18%) for SY compared to SD, respectively. The number of patients in each group with low (51% versus 50%), intermediate (41% each) or high (8% versus 9%) RS was similar in the SD compared to SY groups, respectively. There was a statistically significant association between RS and tumour grade (P=<0.001) and lymphovascular invasion (P=<0.001). We did not observe a statistically significant association between RS and MOD (P=0.086). Conclusions: In patients with clinically intermediate breast cancer sent for OncotypeDX testing we found no association between RS and MOD.