Reliability of ¹⁸F-FDG PET Metabolic Parameters Derived Using Simultaneous PET/MRI and Correlation With Prognostic Factors of Invasive Ductal Carcinoma: A Feasibility Study.

Abstract

The objective of our study was to correlate semiquantitative PET parameters-standardized uptake value (SUV) and total lesion glycolysis (TLG)-derived in simultaneous PET/MRI using MRI-based attenuation correction with clinical and histopathologic prognostic factors in patients with breast cancer. Eighty-two invasive ductal carcinomas in 69 women were included in the study. All the subjects underwent whole-body (WB) PET/MRI (supine WB mode) and dedicated PET/MRI of the breast (prone breast imaging mode) for staging on a simultaneous PET/MRI system. The SUV and TLG values were calculated from ¹⁸F-FDG PET data using MRI-based attenuation correction (2-point Dixon sequence for tissue segmentation). Relationships between SUV and TLG values and clinical and histopathologic parameters (i.e., tumor size, tumor grade, Ki-67 status, and hormonal receptor expression status) were evaluated using Spearman correlation coefficient analysis. A significant correlation was observed between mean SUV (SUV_mean) and maximum SUV (SUV_max) values derived with WB PET and regional PET of the breasts performed simultaneously with MRI (r = 0.88 and 0.89, respectively). A significant difference (p < 0.05) was observed in SUV_mean, SUV_max, and TLG values between the grades and molecular subtypes of breast cancer. High SUV_mean, SUV_max, and TLG values were found to correlate with larger tumor size (p < 0.01), higher proliferation index (p < 0.05), higher grade (p < 0.01), and triple-negative hormonal receptor status (p < 0.01, p < 0.05). Semiquantitative FDG parameters derived with MRI-based attenuation correction in simultaneous PET/MRI are reliable and correlate with clinicopathologic features such as grade as well as subtype and thus could be used in the prognostication of breast cancer.