F-18 FDG PET/MRI IN PRIMARY STAGING OF BREAST CANCER: THE RELATIONSHIPS BETWEEN IMAGING PARAMETERS OF PRIMARY TUMOR AND DISTANT METASTASES

554 Introduction: The aim of this study was to evaluate the relationship between the types of distant metastatic spread and the imaging features of primary tumor on PET/MRI for primary staging in newly diagnosed breast cancer-invasive ductal carcinoma (IDC) patients. Methods: The imaging data of 260 female patients with histopathologically diagnosed IDC who underwent whole body and breast dedicated 18F-FDG PET/MRI for primary staging between 2016-2019 in our department were evaluated retrospectively. Maximum and mean standardized uptake values (SUV), metabolic tumor volume (MTV) and total lesion glycolysis (TLG), Tumor/Healthy Breast Tissue Activity Ratio (THBR), tumor diameter and minimum apparent diffusion coefficient (ADCmin) values of primary tumors were obtained from PET/MR images. The patients were grouped as nonmetastatic (NM), oligometastatic (OM) (1-5 metastatic lesions) and multimetastatic (MM) (>5 metastatic lesions) disease according to the number of distant metastases. The patients with metastasis were also divided into two groups as isolated bone metastasis (IBM) and mixed/soft tissue metastasis (M-SM) groups according to the sites of metastatic spread. Finally, the cases were divided into two groups as “PET positive” and “PET negative” according to the presence of pathologic FDG uptake outside the breast and axillary lymph nodes (LN). The histopathological data and pretreatment CA 15-3 levels of patients were also recorded. Statistical analyses were performed on SPSS version 23.0. Results: The characteristics of patients were shown in Table 1. SUVmax, SUVmean, MTV, TLG, THBR, ADC values and diameters of primary breast tumor had significant differences among metastatic groups (Table 2). For the comparison between OM and MM groups, it was found that tumor diameters only demonstrated significant differences between two groups (p = 0.04). THBR, MTV, TLG and diameter of primary tumors were significantly higher in the M-SM group compared to the IBM group (Table 3). CA 15-3 levels were found to be significantly higher in the MM group compared to the OM group ( 56.0 vs 14.6 median values; p < 0.001) and in the M-SM group compared to the IBM group (38.75 vs 14.9 median values; p = 0003). In logistic regression analysis to predict the “FDG PET positivity”, a regression model with tumor diameter, CA 15-3 levels and molecular subtypes had an accuracy of 82.5% (Table 4). In ROC analysis, tumor diameter and CA 15-3 levels had AUC values of 0.799 and 0.728 respectively. While there were no patients with FDG PET positivity in axillary LN negative group, PET positivity was observed in 44.2% of patients with positive axillary LN (Fisher’s exact test 2-sided p<0.001). Conclusions: Primary tumors of patients with IDC had higher metabolic and glycolytic activity, higher cellularity and larger sizes in multimetastatic state compared to the nonmetastatic and oligometastatic groups and in M-SM group compared to the IBM group. Tumor diameter, pretreatment CA 15-3 levels and molecular subtypes of tumors were found as significant predictors of FDG PET positivity for distant metastasis and extra-axillary LNs.