Assessing the Diagnostic Value of MRI T1rho Mapping in Predicting Molecular Prognostic Biomarkers and Subtypes of Breast Cancer

ABSTRACT Background Prognostic factors and molecular subtypes are important in treatment planning and predicting response to therapy in breast cancer, and the exploration of noninvasive imaging methods to characterize breast cancer has been ongoing. Purpose To evaluate the use of T1rho mapping in predicting the status of prognostic biomarkers and molecular subtypes of breast cancer. Study Type Prospective. Subjects Ninety One women with breast tumors. Field Strength/Sequence 3T, T1 rho prepared balanced turbo field echo sequence. Assessment Mean T1rho values were compared between positive and negative status of prognostic biomarkers [estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), and proliferation index (Ki‐67)] and molecular subtypes. Statistical Tests Independent t test, one‐way analysis of variance, Kruskal–Wallis test, chi‐square test, Fisher exact test, and receiver operating characteristic (ROC) analysis. Results Mean T1rho values were significantly higher in ER‐negative compared to ER‐positive tumors (70 ± 6.769 vs. 55 ± 10.791 ms), in PR‐negative compared to PR‐positive tumors (68.48 ± 9.563 vs. 60.46 ± 10.099 ms), and in high Ki‐67 proliferation compared to low Ki‐67 proliferation tumors (66.59 ± 8.994 vs. 57.77 ± 11.501 ms). Significant negative correlations were observed between ER and PR statuses and T1rho values ( r s = 0.416 and 0.392, respectively). The Ki‐67 status was significantly positively correlated with T1rho values ( r s = 0.369). The Luminal A subtype had a significantly lower T1rho value than other subtypes (56.46 ± 10.553 vs. 66.58 ± 9.204 ms). The Luminal B subtype had significantly lower T1rho values than the triple‐negative (TN) subtype (63.957 ± 9.794 vs. 72.237 ± 8.229 ms). The TN subtype had a significantly higher T1rho value than luminal subtypes (72.237 ± 8.229 vs. 62.903 ± 10.289 ms). The T1rho values had good diagnostic performance in identifying Luminal A and TN breast cancers with areas under the ROC curve of 0.767 and 0.776. Data Conclusion T1rho mapping has the potential to be a non‐invasive imaging biomarker for evaluating the prognostic biomarkers and molecular subtypes of breast cancer.