Pharmacokinetic Tumor Heterogeneity as a Prognostic Biomarker for Classifying Breast Cancer Recurrence Risk

Pharmacokinetic Tumor Heterogeneity as a Prognostic Biomarker for Classifying Breast Cancer Recurrence Risk 555 234 IEEE Transactions on Biomedical Engineering (TBME)

Majid Mahrooghy, Ahmed B. Ashraf, Dania Daye, Elizabeth S. McDonald, Mark Rosen, Carolyn Mies, Michael Feldman, and Despina Kontos, University of Pennsylvania
Volume: 62 , Issue: 6 Page(s): 1585 – 1594

pharmacokinetic-tumor

Breast tumors have been shown to be highly heterogeneous lesions. Intra-tumor heterogeneity presents a major challenge for determining breast cancer prognosis and can affect response to therapy. Histologic measures have classically been used to measure heterogeneity, although a reliable non-invasive measurement is needed both to establish baseline risk of recurrence and monitor response to treatment. Most studies to date have used aggregate measures for tumor characterization, while fewer have focused on capturing heterogeneity using imaging features. Here we propose using spatiotemporal wavelet kinetic features from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) to quantify intra-tumor heterogeneity in breast cancer. Tumor pixels are first partitioned into homogeneous sub-regions using pharmacokinetic measures. Heterogeneity wavelet kinetic (HetWave) features are then extracted from these partitions to obtain spatiotemporal patterns of the wavelet coefficients and the contrast agent uptake. The HetWave features are evaluated in terms of their prognostic value using a logistic regression classifier with genetic algorithm wrapper-based feature selection to classify breast cancer recurrence risk as determined by a validated gene expression assay. Receiver operating characteristic (ROC) analysis and area under the curve (AUC) are computed to assess classifier performance using leave-one-out (LOO) cross validation. The HetWave features outperform other commonly used features (AUC=0.88 HetWave vs. 0.70 standard features). The combination of HetWave and standard features further increases classifier performance (AUCs 0.94). The rate of the spatial frequency pattern over the pharmacokinetic partitions can provide valuable prognostic information. HetWave could be a powerful feature extraction approach for characterizing tumor heterogeneity, providing valuable prognostic information.