557 Background: Oncotype DX Recurrence Score [RS] is used to predict the benefits of chemotherapy added to adjuvant hormone therapy in ER positive early stage breast cancer. It is also prognostic. However, its expense may be a concern in some health care systems and communities. In addition, it is labor intensive, requiring shipment of tissue samples to a single laboratory with an average 10-14 day turnaround time (in the US). A reliable and inexpensive estimator of the Oncotype DX risk score using readily available pathologic variables from tumor specimens could be useful. Methods: We reviewed our prospective database of patients with Oncotype DX results obtained over 2.5 years (1155 specimens with Oncotype DX scores, September 2008 – March 2011) and identified 766 invasive ductal carcinomas with known ER status, PR status, histologic grade, and nuclear grade. Through linear regression analysis, we predicted recurrence score for each tumor using these four parameters. After categorizing according to the same risk classification as in Oncotype DX (low risk: RS<18, intermediate risk: RS 18-30, or high risk: RS>30) we compared our predicted recurrence score to actual Oncotype DX recurrence score. Results: Overall 69.7% of specimens were assigned the same risk category as with the Oncotype DX level. In the predicted low risk group, 1.7% of patients were actually high risk. None of the predicted high risk patients had a low score. Conclusions: We found a strong correlation between scores estimated using our model and actual reported Recurrence Scores. If validated, our model could provide a clinically useful estimation of risk at lower cost. [Table: see text]
A study of association of Oncotype DX recurrence score with DCE-MRI characteristics using multivariate machine learning models
