Novel imaging marker for low- and mid-recurrence score patients at risk for recurrence



The 21-gene (Oncotype DX) recurrence score (RS) has been utilized as a prognostic assay in patients with hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2) breast cancer. The results stratify patients into low-, mid- and high-risk of disease recurrence. Patients with a low recurrence score tend to have lower recurrence rates and are less responsive to chemotherapy. On the other hand, patients with high-recurrence scores are more likely to respond to chemotherapy. While for patients at the high end of the RS spectrum it is apparent that chemotherapy carries benefit, distinguishing chemotherapy benefit for low- and mid-recurrence score (RS-low/-mid) patients is less clear.


To address this, we used a novel 2-paramenter pharmacokinetic modeling framework that allows biosignatures to be extracted from dynamic contrast enhanced (DCE) magnetic resonance imaging (MRI) studies that contain 3-6 timepoints spaced 60-90 seconds apart. These parameters, referred to as P1 and P2, represent leakiness from vessels to the extravascular space and vice versa.


Using this model, we performed a study in a cohort of 111 breast cancer patients with 21-gene assay RS   and DCE-MRIs available. P1 and P2 were extracted from patient standard clinical DCE-MRIs. We performed univariate (Kaplan-Meier) survival analyses to test the prognostic value of the parameters covariate in predicting recurrence free survival (RFS). A pre-specified value of the median P1 and P2 across the patient population was used for stratifying patients into “low” and “high” groups, representing patients with “low” or “high” vessel leakiness. We analyzed patients according to RS. Low P1 showed better outcomes in RS-low and RS-mid patients (n=88, p≤0.028; log-rank test). Patients with a high P1 had a 20.2% chance recurrence at six years. The same trend was observed when considering just RS-mid patients only (n=23, p≤0.058). No recurrences were observed in patients with low P1 in either the RS-low or RS-mid categories. Additionally, there were no recurrences in the high P1, RS-low/-mid category that received chemotherapy, suggesting that chemotherapy could be beneficial in this category of patients, although the trend was not statistically significant (n=10, p=0.46). Similar results were seen for P2.


Overall, we demonstrate how our prognostic imaging marker can identify a subset of patients with low-/mid-recurrence scores that could benefit from chemotherapy. This new prognostic approach has the potential to optimize treatment and improve personalized care by providing clarity on the use of chemotherapy in low-/mid-recurrence score patients.

View the full publication here.