Prediction of tumor differentiation using sequential PET/CT and MRI in patients with breast cancer

Objective The aim of this study is to assess tumor differentiation using parameters from sequential positron emission tomography/computed tomography (PET/CT) and magnetic resonance imaging (MRI) in patients with breast cancer. Methods This retrospective study included 78 patients with breast cancer. All patients underwent sequential PET/CT and MRI. For fluorodeoxyglucose (FDG)-PET image analysis, the maximum standardized uptake value (SUV max ) of FDG was assessed at both 1 and 2 h and metabolic tumor volume (MTV) and total lesion glycolysis (TLG). The kinetic analysis of dynamic contrast-enhanced MRI parameters was performed using dynamic enhancement curves. We assessed diffusion-weighted imaging (DWI)–MRI parameters regarding apparent diffusion coefficient (ADC) values. Histologic grades 1 and 2 were classified as low-grade, and grade 3 as high-grade tumor. Results Forty-five lesions of 78 patients were classified as histologic grade 3, while 26 and 7 lesions were grade 2 and grade 1, respectively. Patients with high-grade tumors showed significantly lower ADC-mean values than patients with low-grade tumors (0.99 ± 0.19 vs.1.12 ± 0.32, p = 0.007). With respect to SUV max 1, MTV2.5, and TLG2.5, patients with high-grade tumors showed higher values than patients with low-grade tumors: SUV max 1 (7.92 ± 4.5 vs.6.19 ± 3.05, p = 0.099), MTV2.5 (7.90 ± 9.32 vs.4.38 ± 5.10, p = 0.095), and TLG2.5 (40.83 ± 59.17 vs.19.66 ± 26.08, p = 0.082). However, other parameters did not reveal significant differences between low-grade and high-grade malignancies. In receiver-operating characteristic (ROC) curve analysis, ADC-mean values showed the highest area under the curve of 0.681 (95%CI 0.566–0.782) for assessing high-grade malignancy. Conclusions Lower ADC-mean values may predict the poor differentiation of breast cancer among diverse PET–MRI functional parameters.