The Relationship between Histological Composition and Metabolic Profile in Breast Tumors and Peritumoral Tissue Determined with (1)H HR-MAS NMR Spectroscopy

SIMPLE SUMMARY: Breast cancer is the most prevalent malignancy all over the world. Intraoperative histological imaging of frozen tissue sections is frequently used in the evaluation of surgical margins for the presence of residual transformed cells. However, this method is time consuming, prone to sampling error and subjective interpretation. The purpose of this work was to investigate a sensitivity of the (1)H HR-MAS NMR (high-resolution magic angle spinning nuclear magnetic resonance) technique in the detection of cancer cells in the tissue material collected during breast-conserving surgery and to gain an insight into a metabolic reprogramming of intratumoral fibrosis. The multivariate classification models permitted discrimination of the cancerous from non-cancerous samples with an accuracy of 87%. The accumulation of several metabolites (inter alia lactate, glutamate, succinate) in the fibrotic tissue within the tumor in reference to the extratumoral fibrous connective tissue was detected. The results of our work contribute to the increased understanding of breast cancer heterogeneity. ABSTRACT: Breast tumors constitute the complex entities composed of cancer cells and stromal components. The compositional heterogeneity should be taken into account in bulk tissue metabolomics studies. The aim of this work was to find the relation between the histological content and (1)H HR-MAS (high-resolution magic angle spinning nuclear magnetic resonance) metabolic profiles of the tissue samples excised from the breast tumors and the peritumoral areas in 39 patients diagnosed with invasive breast carcinoma. The total number of the histologically verified specimens was 140. The classification accuracy of the OPLS-DA (Orthogonal Partial Least Squares Discriminant Analysis) model differentiating the cancerous from non-involved samples was 87% (sensitivity of 72.2%, specificity of 92.3%). The metabolic contents of the epithelial and stromal compartments were determined from a linear regression analysis of the levels of the evaluated compounds against the cancer cell fraction in 39 samples composed mainly of cancer cells and intratumoral fibrosis. The correlation coefficients between the levels of several metabolites and a tumor purity were found to be dependent on the tumor grade (I vs II/III). The comparison of the levels of the metabolites in the intratumoral fibrosis (obtained from the extrapolation of the regression lines to 0% cancer content) to those levels in the fibrous connective tissue beyond the tumors revealed a profound metabolic reprogramming in the former tissue. The joint analysis of the metabolic profiles of the stromal and epithelial compartments in the breast tumors contributes to the increased understanding of breast cancer biology.