Magnetic Resonance Spectroscopy in the Diagnosis and Prognosis of Breast Cancer

Abstract

Objective. To evaluate the efficiency of magnetic resonance spectroscopy in the diagnosis and prognosis of breast cancer (BC).Material and methods. Twenty-eight patients aged 37 to 80 years with established primary invasive BC were examined. Its grade was determined according to pathomorphological verification with immunohistochemical analysis. Breast MRI was performed using the standard protocol, by determining the measured diffusion coefficient (MDC), dynamic contrast enhancement (DCE). Proton magnetic resonance spectroscopy was carried out using the Breeze software package. The clinical and morphological findings and the results of radiation studies were compared to determine Nottingham prognostic index (NPI) scores. To identify the Spearman rank correlation coefficient between MRI findings (the type of pharmacokinetic curves, the total choline-containing (tCho) peak integral) and the values characterizing tumor size and grade.Results. NPI calculation showed that the scoring range was 2.4 to 6.76; the expected 5-year survival rates reached 93%. The mean MDC was 0.856×10-3 mm2/sec; type II pharmacokinetic curve prevailed (n=16; 57.1%). There was a statistically significant negative correlation between the values of MDC, the type of a contrast agent accumulation curve, the levels of HER2/neu and the proliferation marker Ki-67; there was a statistically significant strong positive correlation between the presence of 1H-MPC tCho peak and the indicators determining tumor malignancy (the levels of HER2/neu and Ki-67). Statistically significant differences between the type of a pharmacokinetic curve in DCE and the level of tCho in the prognostic groups defined when calculating NPI were determined in patients with moderate (n=4; 14.2%) and poor (n=10; 35.7%) prognosis.Conclusion. The capabilities of MR spectroscopy are superior in information content to the data obtained by determining the MDC, the nature of contrast medium accumulation in breast tumor, and are comparable with the data determining the type of a tumor (the presence of the HER2/neu gene, Ki-67 proliferation marker) in predicting BC cancer grade and 5-year survival rates.