The impact of FDG PET in the staging of breast cancer

Abstract

530 Background: To evaluate the impact of whole-body fluorine-18-fluorodeoxyglucose (FDG) positron emission tomography (PET) in staging breast cancer beyond the breast and axilla. Methods: 200 women, mean age 51 yrs (range 28–81) with breast cancer were enrolled in an IRB approved multi-modality imaging trial from 3/02 to 5/05. After imaging with film screen mammography ± ultrasound (US), patients underwent digital mammography, whole breast US, magnetic resonance imaging (MRI), and whole body FDG PET. Imaging reports and patient charts were reviewed and distant lesions were classified: 1 = no uptake; 2 = physiologic uptake; 3 = non-suspicious uptake; 4 = suspicious uptake without work-up; 5 = suspicious uptake with work-up. Decisions to forego work-up in class 4 lesions were clinical. Class 5 PET lesions were evaluated by confirmatory studies, including computed tomography (CT), MRI, x-ray (XR), bone scan (BS), or pathology. Patients also underwent standard staging with chest XR and BS per clinical judgment. Results: 189/200 (95%) women received PET scans. PET identified 71 distant areas of increased uptake in 42/189 (22%) women. 59/71 (83%) lesions were considered suspicious in 33/42 (79%) women: 50/59 (85%) lesions were class 5 and 9/59 (15%) class 4. Of the 50 class 5 lesions, 19 (38%) were true positive (TP) for neoplastic disease, 2 (4%) were TP for non-neoplastic disease, and 29 (58%) were false positive (FP). The positive predictive value (PPV) for class 5 lesions was 40%. The 19 TP distant sites of malignancy occurred in 6/189 (3%) patients; 4 of these patients (67%) also had TP distant sites identified by standard staging procedures. PET alone identified TP distant disease in 2/189 (1%) women. PET identified 29 FP findings in 22/189 (12%) patients. FP findings prompted 17 CTs, 8 XRs, 2 MRIs, and 1 BS. MRI revealed 2 additional neoplasms that had been false negatives (FN) by PET in 2/189 (1%) women. Conclusions: Whole-body FDG-PET added little additional information in staging our patients with known primary breast cancer, based on low PPV and equivalent TP and FN rates. Additionally, the high FP rate of PET in these patients may prompt unnecessary, costly, and invasive follow-up studies. No significant financial relationships to disclose.