Prediction of Indolent Breast Cancer with Favorable Prognostic Factors by Metabolic Profiling Using In Vivo and Ex Vivo MR Metabolomics

Abstract

To evaluate whether metabolic profiles obtained using high-resolution magic angle spinning magnetic resonance spectroscopy (HR MAS MRS) and total choline-containing compound (tCho) on in vivo MRS could predict indolent tumors based on highly favorable prognostic factors. We analyzed 50 frozen tissue samples from 50 patients (mean 46.4 years, range 29–72 years) with breast cancer using HR MAS MRS. In vivo single-voxel MRS analyses were also performed on these patients preoperatively. We defined estrogen receptor (ER)-positive tumors with a low histological grade, low Ki-67-positivity (<14 %), and negative lymph node metastases as an indolent tumor. By univariate analysis, metabolic profiles on HR MAS MRS and tCho on in vivo MRS were compared according to dichotomized pathological parameters using the Mann–Whitney test. By multivariate analysis, orthogonal projections to latent structure-discriminant analysis (OPLS-DA) were performed to differentiate groups with different prognostic pathological parameters. A total of 6 indolent tumors (12 %) and 44 non-indolent tumors (88 %) were studied. By univariate analysis, tumors without recurrence showed significantly higher Tau and Cr values than those with recurrence (P = 0.041, respectively). By multivariate analysis, an OPLS-DA model showed sensitivities of 100, 77, and 82 % and specificities of 68, 100, and 96 % for the prediction of indolent tumors, tumors with recurrence, and tumors with lymph node metastases, respectively. By univariate analysis of in vivo MRS, tumors without recurrence showed significantly higher values of tCho than those with recurrence (P = 0.043 and 0.035). Several metabolites of Gly, Lac, Tau, Cr, GPC, and Cho on HR MAS MRS could be potential candidate biomarkers for predicting indolent tumors, tumors with early recurrence, and lymph node metastases. Metabolite profiling using HR MAS MRS might enable the prediction of breast cancer prognoses, especially for ER-positive tumors.