Magnetic resonance imaging (MRI) phenotypes showed promise in predicting breast cancer pathologic stage and lymph node status, according to an article published online ahead of print in Cancer.1

Investigators sought to demonstrate that the use of MRI could accurately predict pathologic stage on computer-extracted image phenotypes (CEIPs) of biopsy-proven breast cancer.

They looked at a data set of 91 breast cancers of deidentified breast MRIs from patients who had information regarding pathologic stage (stage 1, n = 22; stage 2, n = 58; stage 3, n = 11) as well as surgically verified lymph node status (negative lymph nodes, n = 46; ≥ 1 positive lymph node, n = 44; no lymph nodes examined, n = 1). Tumors were characterized according to size and CEIP. Investigators built models that combined 2 CEIPs to predict pathologic stage and lymph node involvement.

Continue Reading

Results showed that tumor size was the best predictor of pathologic stage; however CEIPs that captured biologic behavior were also predictors.

RELATED: Blood Test Aids in Identifying Treatment Resistance in Metastatic Breast Cancer

For example, stage 1 and 2 vs stage 3 demonstrated an area under the curve of 0.83. Tumor size was unsuccessful in predicting positive lymph nodes. Investigators noted that the addition of CEIP that described tumor “homogeneity” improved prediction compared with chance (AUC = 0.62; P = .003).


  1. Burnside ES, Drukker K, Li H, et al. Using computer-extracted image phenotypes from tumors on breast magnetic resonance imaging to predict breast cancer pathologic stage [published online ahead of print November 30, 2015]. Cancer. doi: 10.1002/cncr.29791.