Katerina Dodelzon MD, Presenter: Nothing to Disclose

Jennifer Chun MPH, Abstract Co-Author: Nothing to Disclose

Shira Schwartz, Abstract Co-Author: Nothing to Disclose

Sidney Law BS, Abstract Co-Author: Nothing to Disclose

Elizabeth Kern MPH, Abstract Co-Author: Nothing to Disclose

Freya Schnabel MD, Abstract Co-Author: Nothing to Disclose

Linda Moy MD, Abstract Co-Author: Nothing to Disclose

Oncotype DX breast cancer 21 gene assay recurrence score (RS) is used to predict disease recurrence and response to chemotherapy in order to offer patients the highest treatment benefit to risk ratio. The purpose of this study was to assess whether magnetic resonance imaging (MRI) features can be used to predict Oncotype RS in patients with newly diagnosed, estrogen receptor (ER) positive invasive breast cancer.

This was an IRB-approved HIPAA-compliant retrospective review of 727 women with newly diagnosed invasive breast cancer that were ER positive. Of 727 women, 59 had both Oncotype testing and preoperative MRI at our institution. Variables of interest included age, menopausal status, tumor characteristics, Oncotype RS, mammographic breast density, background parenchymal enhancement (BPE), and other MR imaging characteristics (lesion type, shape, mass enhancement, diameter, volume, presence on first post-contrast imaging, and contrast enhancement kinetics). Statistical analyses were performed using ANOVA and linear regression.

Of 59 patients, the majority had invasive ductal carcinoma (74%) and were Stage I and II (98%), progesterone receptor (PR) positive (88%) and Her2Neu negative ( 93%). 37 of 59 patients (63%) had pre-menopausal status which was significantly associated with a lower oncotype score ( p=0.05). Patients with higher Oncotype RS were significantly more likely to have IDC on histology (p= 0.04). When looking at the MRI tumor characteristics, T2 hyperintensity was significantly associated with higher Oncotype RS (p=0.01). Other MRI features, including tumor size, morphology, and dynamic kinetic enhancement patterns were not significantly associated with Oncotype RS. As expected, Ki-67 proliferation index was statistically associated with Oncotype RS (p=0.0009).

Our study did not support the findings of previous work that have looked at Oncotype RS and MRI features. We found that T2 hyperintensity may play a role in predicting tumor Oncotype RS. Morphology, size and enhancement characteristics do not reliably predict tumor recurrence or response to chemotherapy.

Only tumor T2 hyperintensity may have utility in predicting tumor recurrence and response to chemotherapy in patients diagnosed with ER positive breast cancer.

Dodelzon, K, Chun, J, Schwartz, S, Law, S, Kern, E, Schnabel, F, Moy, L, Correlation between Magnetic Resonance Imaging Characteristics and Breast Cancer Oncotype DX Score.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14010725.html